- Email: [email protected]
A Risk-Based Approach to Establish Stability Testing Conditions for Tropical Countries
COMMENTARY A Risk-Based Approach to Establish Stability Testing Conditions for Tropical Countries MANUEL ZAHN,1 PER W. KA˚LLBERG,2 GIJSBERT M. SLAPPENDEL,3 HAROLD M. SMEENGE3 1
AstraZeneca R&D So¨derta¨lje, Regulatory CMC, So¨derta¨lje, Sweden
2
European Centre for Medium-Range Weather Forecasts, Reading RG2 9AX, United Kingdom
3
AstraZeneca R&D So¨derta¨lje, So¨derta¨lje, Sweden
Received 19 April 2005; revised 24 November 2005; accepted 21 December 2005 Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20584
ABSTRACT: The external stability risk factors heat and moisture are evaluated with respect to the development of pharmaceutical products intended to be marketed in tropical and subtropical countries. The mean daily temperatures and dewpoints measured four times per day at selected places in Southeast Asia, South America, China, Southern Africa and the Caribbean are used to calculate the daily and monthly fluctuations of temperature and partial water vapour pressure, the mean kinetic temperature and the relative humidity. Based on these data, the hottest and the most humid place in each country or region are identified to reflect the worst case for the specific region. A formula to calculate safety margins for temperature and partial vapour pressure is introduced taking into consideration the difference between measured meteorological parameters and the stability testing conditions. An appropriate long-term stability testing condition is proposed for each selected country, related to the worst case for each specific region and the safety margins, as well as its classification in either Climatic Zone IVA or IVB. ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 95:946–965, 2006
Keywords: physicochemical properties; tropical climate; hot and humid; high humidity; testing condition; safety margin; climatic zone; thermodynamics; stability; regulatory science
INTRODUCTION In more than 30 years of ongoing discussions, pharmaceutical manufacturers could not agree with regulators on a common long-term stability All web addresses in References have been accessed in March 2006. Gijsbert M. Slappendel is a Pharmacy student placed for internship at Regulatory CMC during December 2004 to February 2005 and Harold M. Smeenge is a Pharmacy student placed for internship in 2004 at Regulatory Affairs, So¨derta¨lje, Sweden. Correspondence to: Manuel Zahn (Telephone: þ46-8-552 57640; Fax: þ46-8-552 54660; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 95, 946–965 (2006) ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association
946
testing condition, which would be adequate for hot and humid countries. While testing at 308C/70% RH became common practice in industry, and was also recommended by the World Health Organization (WHO)1 in 1996, in 2001 the WHO changed the recommended testing condition for hot and humid countries to 308C/65% RH2. Subsequently, the International Conference on Harmonisation (ICH) Stability Guideline Q1F recommended the same condition3. Regulators representing the Association of South East Asian Nations (ASEAN), however, decided in 2004 not to follow this advice, and required in their published guideline4 to test at 308C/75% RH instead. In December 2004, the WHO encouraged5 manufacturers and regulators
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
Table 1.
Criteria used to Classify Climatic Zones and Testing Conditions
CZ
Definition
I II
Temperate climate Subtropical and Mediterranean climate Hot and dry climate Hot and humid climate
III IV
947
Criteria, Mean Annual Temperature Measured in the Open Air/Mean Annual Partial Water Vapour Pressure
Testing Conditions [8C/% RH]
158C/11 hPa >15 to 228C/>11 to 18 hPa
21/45 25/60
>228C/15 hPa >228C/>15 hPa
30/35 30/70
to propose a solution, which could be acceptable to all parties involved. In August 2005, the Brazilian authorities introduced 308C/75% RH as the longterm condition required for Brazil6. The 40th WHO Expert Committee on Specifications for Pharmaceutical Preparations met in Geneva in October 20057, and decided to split the current Climatic Zone IV (hot and humid countries) into CZ IVA, for which 308C/65% RH will remain the standard long-term testing condition, and CZ IVB, for which, if justified, 308C/75% RH will become the testing condition. In the following, the climate of a selected number of hot and humid countries in Climatic Zone IV are presented and analysed, and the most appropriate testing conditions for long-term stability studies are proposed to facilitate the categorisation of different hot and humid countries in Climatic Zone IVA and IVB.
CLIMATIC ZONES
19728 and Wolfgang Grimm in 19869 and in 199810 (see Tab. 1) when they defined four different long-term test conditions, which match with the climatic conditions of the target markets categorised in just four different climatic zones. This concept is described in regulatory guidelines11 and pharmacopoeias12 and became an established standard in developing pharmaceutical products. As a result of the above-mentioned WHO proposal to split the Climatic Zone IV into two different zones, IVA and IVB, there is a need for criteria to categorize a country to one of these CZs. We suggest to apply the criteria listed in Table 2. Ko¨ppen It is proposed to use the Ko¨ppen classification of world climates in order to be able to distinguish between different regional climates in Climatic Zone IV countries as being either ‘‘Moist tropical’’ (Group Af, Am and Aw) or ‘‘Humid subtropical’’ (Cfa).
Schumacher/Grimm In order to be able to reduce the amount of stability testing, the number of different testing conditions must be reduced to a sufficient extent. This has been proposed by Paul Schumacher in Table 2.
DOI 10.1002/jps
CALCULATION OF SAFETY MARGINS For temperature, the following equation can be applied to calculate a safety margin:
Proposal for Criteria to Split Climatic Zone IV
CZ
Definition
Criteria, Mean Annual Temperature Measured in The Open Air/Mean Annual Partial Water Vapour Pressure
IVA IVB
Hot and humid climate Hot and very humid climate
>228C/27 hPa >228C/>15 to 27 hPa
Testing Conditions [8C/% RH] 30/65 30/75
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
948
ZAHN ET AL.
YT ¼ ðTS TÞ 100=T YT, margin of safety for temperature [%]; TS, stability testing storage temperature; T, temperature measured in the environment or calculated as MKT. YT > 0
if
TS > T
YT ¼ 0
if
TS ¼ T
There is a probability of failure if YT < 0. The same principle can be applied to other parameters like partial water vapour pressure. The amount of safety margin required is dependent on: the impact of the environment on the product during shelf-life (variability in loading), and on the other hand on the strength of a particular batch of a product to resist heat and moisture (‘‘variability in resistance’’). The distributions of the loading and the resistance result in the ‘‘probability of failure’’ of a particular product in a particular market. The variability in resistance can be neglected, as the batch-to-batch variability of a product’s stability is too low to be taken into consideration. As a consequence, the focus is on the variability in loading, for example, the fluctuations of heat and moisture in the environment or the frequency of extreme temperatures and humidities during the shelf life of a particular product.
CLIMATIC DATA
Figure 1. Daily temperature fluctuations—El Nido, The Philippines.
means at 00UTC, 06UTC, 12UTC and 18UTC. ‘Screen level’ is normally about 2 m above ground where the standard meteorological instruments are placed. Vapour pressures—applying an updated version of Wexler’s equation14—and relative humidities were then determined from analysed temperatures and dewpoints using basic thermodynamic relationships. The additional parameter mean kinetic temperature (MKT), which includes the reaction rate constants in the evaluation of heat effects on pharmaceutical products, was calculated by applying an equation derived by Haynes15 based on the Arrhenius equation. The activation energy Ea ¼ 83.144 kJ/ mol has been used. The climatic parameters were then analysed in order to identify: the daily and monthly fluctuations of temperature and vapour pressure; their mean maximum values; the place representing the most loading climatic condition in each country or region.
Quantification of the variability in loading starts with the determination of key meteorological parameters at different times of the day and the year, leading to identification of the most loading part of a country or region. Global climatic data have been extracted from the ERA-40 reanalysis of ECMWF.13 Reanalysis is a process in which global weather observations, near the ground and in the free atmosphere, are assembled to form a regular mesh covering the earth with a resolution of about 125 km and at 60 vertical levels using a sophisticated meteorological forecasting system. Measurements take place every 6th hour. Reanalysed screen level temperatures and dewpoints from 1979 to 2002 were averaged into monthly
Figure 2. Daily partial vapour pressure fluctuations—El Nido, The Philippines.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
949
Table 3. Climatic Data for ASEAN Member Countries
Country
City
Brunei
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
25.5
25.8
16
27.35
83.8
64.4
Cambodia
Phnom Penh
26.8
27.2
11
27.69
78.5
65.2
Indonesia
Jakarta, Java
27.1
27.2
10
28.96
80.6
68.2
Kolbano, Timor
27.6
27.7
8
27.49
74.5
64.7
SW Sulawesi
27.4
27.5
9
28.82
78.9
67.9
Palembang, Sumatra
26.7
26.9
11
29.17
83.2
68.7
Surabaya, Java
27.6
27.7
8
29.07
78.7
68.4
Laos
Vianchan (Vientiane)
24.9
25.6
17
24.22
77.0
57.0
Malaysia
Kuala Lumpur
26.1
26.5
13
27.91
82.7
65.7
Myanmar
Yangon
26.6
27.5
9
25.66
73.5
60.4
Philippines
Cebu, Cebu
27.3
27.4
9
30.18
83.1
71.1
Davao, Mindanao
25.7
25.9
16
28.64
86.7
67.4
El Nido, Palawan
28.0
28.1
7
30.28
80.2
71.3
Manila, Luzon
26.9
27.0
11
29.33
82.9
69.1
Roxas, Panay
27.4
27.6
9
29.10
79.7
68.5
27.2
27.5
9
29.14
80.7
68.6
Singapore
Thailand
Bangkok
27.9
28.4
6
27.17
72.3
64.0
Vietnam
Hanoi
23.8
24.9
21
24.08
81.9
56.7
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70
1 9 17 0 7 15 5 3 10 0 8 16 4 3 11 5 2 9 5 2 10 14 23 32 1 7 14 8 16 24 9 2 6 4 4 11 9 2 5 6 1 9 5 2 10 5 2 9 2 9 17 15 24 (Continued)
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
950
ZAHN ET AL.
Table 3. (Continued)
City
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Ho Chi Minh City (Saigon)
27.3
27.6
9
28.11
77.6
66.2
Country
Testing Condition [8C/% RH]
YPD [%]
30/75 30/65 30/70 30/75
32 2 6 13
T, mean temperature, calculated by using the sum of 48 measured temperatures (4 temperatures per day for each month), divided by 48; MKT, mean kinetic temperature, calculated as described above; YT, safety margin for temperature, calculated using the MKT versus the testing temperature 308C (for details please refer to chapter Calculation of safety margins); PD, mean partial water vapour pressure, calculated by taking the dewpoints; RH, mean relative humidity, calculated by using the saturation vapour pressure PS at the measured temperature, and the value for PD found in the previous column; RH at 308C, mean relative humidity, calculated by using the saturation vapour pressure PS at the testing temperature 308C, and the value for PD found in the previous column; YPD, safety margin for partial vapour pressure, calculated using the meteorological PD value versus the PD value calculated for the respective testing condition found in the previous column.
SELECTION OF COUNTRIES For the purpose of this exercise, the following countries with tropical and subtropical moist climates have been selected for detailed evaluation, that is countries in Group A and Cfa according to Ko¨ppen:
All of the 10 ASEAN member countries, Brazil and other countries in South America, China (Mainland), Hong Kong and Macau, South Africa, and most of the other SADC member countries in southern Africa, Some Caribbean Islands.
the southeast that belong to Group Aw. Kolbano (on the island Timor in the south-west, west of Kupang) is the place with the highest monthly mean temperatures in two consecutive months (October and November) reaching a peak at 31.58C. The highest monthly mean values for PD in Indonesia have been found at the same place: 31.51 hPa in January with values above 31 hPa in four consecutive months (December to March). The Philippines
All of the Indonesian islands belong to the Ko¨ppen Group Af with the exception of some islands in
The Philippines can be characterised as tropical islands with only minor fluctuations of temperatures and partial water vapour pressure during the day or year. Mean maximum temperatures normally do not exceed 318C, and never drop below 238C. PD values can be found in the range of 25.0 to 33.0 hPa. All of the islands belong to Ko¨ppen Group Am; just a small part of the northern island Luzon is characterised by Group Aw.
Figure 3. Daily Brazil.
fluctuations—Pau,
Figure 4. Daily partial vapour pressure fluctuations—llha Macuapanim, Brazil.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
ASEAN Indonesia
temperature
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
951
Table 4. Climatic Data for Brazil
City, Federal State
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Bele´m, Para´
26.7
26.8
12
29.29
83.4
69.0
Brası´lia (capital)
22.9
23.4
28
18.69
67.1
44.0
Fortaleza, Ceara´
27.1
27.1
11
28.49
79.5
67.1
Ilha de Marajo´, Para´
26.5
26.7
13
29.97
86.6
70.6
Ilha Macuapanim, Amazonas
26.0
26.3
14
30.24
89.7
71.2
Macapa´, Amapa´
26.4
26.7
12
29.24
84.9
68.8
Manaus, Amazonas
27.0
27.4
10
28.33
79.2
66.7
Natal, Rio Grande do Norte
26.7
26.8
12
28.00
79.7
65.9
Pau, Rio Grande do Norte
28.3
28.9
4
22.02
57.1
51.8
Poˆrto Alegre, Rio Grande do Sul
19.7
20.4
47
18.33
79.8
43.2
Recife, Pernambuco
25.9
26.1
15
25.56
76.3
60.2
Rio de Janeiro, Rio de Janeiro
22.9
23.3
29
22.41
80.1
52.8
Salvador, Bahia
26.0
26.1
15
26.60
79.1
62.6
Sa˜o Joa˜o, Amapa´
26.8
26.8
12
28.82
81.7
67.9
Sa˜o Luı´s, Maranha˜o
27.0
27.0
11
29.06
81.5
68.4
Sa˜o Marcelino, Amazonas
25.3
25.5
18
29.06
90.1
68.4
Sa˜o Paulo, Sa˜o Paulo
20.6
21.1
42
18.70
77.2
44.0
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65
6 2 9 48 59 70 3 4 12 8 1 6 9
30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65
2 5 6 2 9 3 5 12 1
30/70 30/75 30/65 30/70 30/75 30/65
6 14 25 35 45 51
30/70 30/75 30/65 30/70 30/75 30/65
62 74 8 16 25 23
30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65
33 42 4 12 20 4 3 11 5 2 10 5 2 10 48 (Continued)
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
952
ZAHN ET AL.
Table 4. (Continued)
City, Federal State
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Teresina, Piauı´
27.2
27.6
9
25.77
71.5
60.7
Uraricoera, Roreima
27.1
27.5
9
23.72
66.3
55.9
The most loading place in the Philippines and probably in the whole ASEAN region identified so far is El Nido in the north of the island Palawan with mean temperatures between 25.98C and 31.08C (average per year: 28.08C), and PD values between 27.3 and 33.1 hPa (average per year: 30.3 hPa). As an example of the graphical presentation of climatic parameters, the daily temperatures and partial vapour pressures per month for El Nido (mean values 1979–2001) are shown in Figures 1 and 2.
Testing Condition [8C/% RH]
YPD [%]
30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
59 70 7 15 24 16 25 34
A list of the key climatic parameters measured and calculated for all ASEAN member countries (Tab. 3) facilitates the selection of the most loading place.
calculated at the place with the highest temperature (El Nido, Philippines: 28.18C), or a safety margin of 6% to the maximum MKT found (Bangkok, Thailand: 28.48C), to get a temperature for long-term stability testing of 308C. To calculate an adequate relative humidity for stability testing, the highest mean value for PD (El Nido, Philippines: 30.28 hPa) is kept constant when moving to a testing temperature of 308C, at which the saturation water vapour pressure is 42.47 hPa, which results in a relative humidity of 71.3% (RH ¼ PD 100/PS). Testing at 308C/70% RH would represent the climatic conditions at the most loading place in the Philippines and the ASEAN region. Seventy percent of relative humidity, however, would give a negative safety margin of 2%.
Testing Conditions for ASEAN
BRAZIL
Comparison of Places in ASEAN
With regard to an appropriate testing temperature, a safety margin of 7% is added to the MKT
Near the equator, Brazil belongs to Ko¨ppen Group Af and Am, north and south of the equator to Group Aw, in the south-east to Group Cfa and in between to Group Cw.
Figure 5. Daily temperature fluctuations—Fortin Paredes, Bolivia.
Figure 6. Daily partial vapour pressure fluctuations—Northeast corner, Bolivia.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
953
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
Table 5. Climatic Data for Bolivia
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Cavinas, Beni
25.3
25.7
17
26.57
82.5
62.6
Fortı´n Paredes, Santa Cruz
25.5
26.2
14
21.21
64.9
49.9
Mapiri River, Pando
25.4
25.8
16
27.35
84.0
64.4
Northeast corner, Pando
25.4
25.8
16
27.46
84.6
64.7
The highest temperatures in Brazil have been identified in Pau, Rio Grande do Norte, south of Fortaleza. Four consecutive months (from September to December) show mean maximum temperatures above 34.08C, and no mean minimum temperature below 23.78C is found (see Fig. 3). As a consequence, the adequate temperature for long-term stability testing of medicinal products to be marketed in Brazil would be 308C. That value includes a safety margin of 4% added to the highest MKT calculated for Brazil. The highest values for PD are to be found near the Amazonas river, in particular on the Amazonas island Ilha Macuapanim west of Jau´ National Park, where the PD values never decrease below 27.7 hPa and can go up to 32.5 hPa with a mean of 30.2 hPa (see Fig. 4). This island, however, is not populated in contrast to the other extreme places identified like Bele´m, the capital of Para´, Fortaleza, the capital of Ceara´,
Figure 7. Colombia. DOI 10.1002/jps
Daily temperature fluctuations—Arjona,
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
4 12 20 30 40 50 1 9 17 1 8 16
Macapa´, the capital of Amapa´, Manaus, the capital of Amazonas, Natal, the capital of Rio Grande do Norte and Sa˜o Luı´s, the capital of Maranha˜o.
All of these major cities, however, show a small but positive safety margin for the testing condition 308C/70% RH. In the following, the analysis for all major cities in Brazil, all capitals of federal states in tropical climates, as well as the hottest and most humid places are presented. The safety margin for PD is calculated for testing conditions 308C/65% RH (first line), 308C/70% RH (second line) and 308C/ 75% RH (third line) for each selected place (see Tab. 4).
OTHER COUNTRIES IN SOUTH AMERICA Bolivia The Andes mountains in the southwest of Bolivia are creating a highland climate (Ko¨ppen Group H), while the lowland plains of the Amazon Basin in the northeast are responsible for a humid
Figure 8. Daily partial vapour pressure fluctuations—San Juan, Antioquia, Colombia. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
954
ZAHN ET AL.
Table 6. Climatic Data for Colombia
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Arjona, Bolı´var
27.4
27.5
9
29.87
81.8
70.3
Bogota´ (capital)
19.1
19.4
55
17.50
79.1
41.2
El Banco, Magdalena
26.6
26.8
12
28.84
82.9
67.9
La Pedrera, Amazonas
25.4
25.6
17
28.81
89.0
67.8
Meta, west part
26.0
26.3
14
26.81
79.7
63.1
Sahagu´n, Co´rdoba
26.9
27.1
11
29.53
83.3
69.5
Santa Clara, Amazonas
25.5
25.7
17
28.97
89.0
68.2
San Juan, Antioquia
27.2
27.3
10
29.96
82.9
70.5
Santa Marta, Magdalena
26.7
26.8
12
28.92
82.3
68.1
Sa. de Chiribiquete, Guaviare
25.7
26.0
15
24.86
75.2
58.5
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
8 1 7 58 70 82 4 3 10 4 3 11 3 11 19 7 1 8 5 3 10 8 1 6 5 3 10 11 20 28
tropical wet and dry (savanna) climate (Group Aw). In between, there is a moist climate with mild dry winters (Group C). The hottest place is close to the border to Paraguay in the southeast, east of Fortı´n Paredes,
department Santa Cruz (59.68W, 19.28S) (see Fig. 5). The most humid place identified is in the northeast corner of the country, close to the border to Brazil, in the department Pando (65.38W,
Figure 9. Daily temperature fluctuations—Sarayacu, Pastaza, Ecuador.
Figure 10. Daily partial vapour pressure fluctuations—Yasuni N.P., Orellana, Ecuador.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
955
Table 7. Climatic Data for Ecuador
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Jipijapa, Manabı´
23.7
23.8
26
25.21
86.0
59.4
Sarayacu, Pastaza
26.5
27.0
11
23.69
68.4
55.8
Quito (capital)
18.7
19.1
57
17.25
80.1
40.6
Yasunı´ N.P.
25.2
25.5
18
27.40
85.7
64.5
10.28S). In 6 months from November to April, the mean maximum values for partial pressure exceed 30 hPa (see Fig. 6). West of this corner is another humid place, east of the river Mapiri, south of the border to Brasil (66.48W, 10.28S). The other extreme place assessed is east of Cavinas, department Beni (66.48W, 12.58S) (see Tab. 5).
Colombia Colombia presents a highland climate (Ko¨ppen Group H) stretching from the southwest to the north part of the country, Group BS climate in the north near the coast to the Caribbean Sea, Group Aw in the north and Group Af in the south and following the Pacific coastline in the northwest. Extreme places identified are:
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
10 18 26 17 26 35 60 72 85 1 9 16
Area west of La Chorrera, department Amazonas, in the south; Tarapaca´, located in the tropical rain forest north of Letı´cia, department Amazonas, at the Amazonas river in the south at the border to Brazil and Peru; Sahagu´n, department Co´rdoba; Sa. de Chiribiquete, department Guaviare, in the south.
The hottest place identified in Colombia is in the northwest at 75.48W, 10.08N, south of Arjona, at the coast to the Caribbean Sea southeast of Cartagena, department Bolı´var. Mean temperatures increase to more than 308C in January, February and March, and are never lower than 258C (see Fig. 7). This is also one of the most humid places found in Colombia.
Table 8. Climatic Data for Guyana
City
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Georgetown (capital)
25.4
25.6
17
28.26
86.9
66.5
Skeldon
25.8
26.0
15
28.58
85.9
67.3
Raad
26.4
26.7
13
25.35
73.7
59.7
DOI 10.1002/jps
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
2 5 13 3 4 12 9 17 26
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
956
ZAHN ET AL.
Figure 11. Daily temperature fluctuations—Puerto Curaray, Peru.
The most humid place in Colombia is in the northwest at 76.68W, 8.98N near San Juan (department Antioquia), at the coast to the Caribbean Sea. The mean maximum partial pressure values increase to more than 31 hPa from April to August, and are always higher than 28 hPa with minimum values in January and February (see Fig. 8). The climatic parameters for Colombia are summarised in Table 6.
Figure 12. Daily partial vapour pressure fluctuations—Northeast corner, Peru.
Ecuador Group Aw climate can be found along the coast to the Pacific Ocean in the west, and Group Af in the east part of the country. The capital Quito offers a moderate climate due to the fact that the city is located at 2,800 m above sea level. Mean temperatures fluctuate between 158C and 258C, and values for partial vapour pressure never increase to more than 20 hPa.
Table 9. Climatic Data for Peru
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Lima (capital)
15.2
16.2
85
11.34
65.7
26.7
Nauta, Loreto
25.9
26.3
14
28.14
84.3
66.3
Northeast corner, Loreto
25.4
25.7
17
28.61
87.9
67.4
Puerto Curaray, Loreto
26.5
27.0
11
25.73
74.2
60.6
Puerto Salvatierra, Loreto
26.5
26.9
12
25.05
72.4
59.0
River Tamaya, Ucayali
24.8
25.1
20
27.54
88.2
64.8
San Antonio, Loreto
26.5
26.9
11
26.16
75.5
61.6
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
143 162 181 2 6 13 4 4 11 7 16 24 10 19 27 0 8 16 6 14 22
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
The highest mean temperatures have been found near Sarayacu, province Pastaza, east of Riobamba. Temperatures here fluctuate between 228C and 328C, and there is not a single day throughout the year when the mean maximum temperature drops below 308C (see Fig. 9). The most humid place in Ecuador is the Yasunı´ National Park, province Orellana, in the east close to the border to Peru (see Fig. 10). The place with the highest mean relative humidity is Jipijapa, province Manabı´, south of Manta at the east coast to the Pacific. The mean temperatures fluctuate by not more than 28C between day and night, but there is a difference between summer and winter: The mean minimum temperatures can be detected at night in August at just above 218C, and the mean maximum temperatures during the day in March at almost 278C (see Tab. 7). Guyana The climate in Guyana is in Group Am. There is a Group Aw climate in the southwest part of the country. The hottest place is the savanna in the southwest of the country, southeast of Raad at 59.68W, 2.28N, with mean maximum temperatures reaching 318C (during the day in March and again in September and October), and mean minimum temperatures above 238C (during the night in July). The most humid place can be found in the northeast part of Guyana near the North Atlantic coast, southwest of the small city Skeldon at 57.38W, 5.68N. The mean partial vapour pressure values never drop below 26 hPa (in February), but reach peaks of more than 30 hPa (in July). Details are listed in Table 8. Peru The western coastal plain facing the Pacific Ocean belongs to Group BW, the Andes in the centre to
Table 10.
Figure 13. Daily temperature fluctuations—Paramaribo, Suriname.
Group H and the Amazon Basin in the northeast part to Group Af. The hottest place in Peru is west of Puerto Curaray, department Loreto (74.38W, 2.48S). The daily mean maximum temperatures never drop below 298C, and the lowest mean temperature found is 22.68C (at night in July) (see Fig. 11). The most humid place has been identified as a place in the northeast corner of Peru, department Loreto, north of the Amazon river close to the border to Colombia (70.98W, 2.48S) (see Fig. 12). The highest mean relative humidity has been found at the river Tamaya in the department Ucayali, at the border to Brazil (73.18W, 9.08S). This is due to high partial vapour pressure values at lower temperatures. Other extreme places analysed are Puerto Salvatierra, department Loreto (76.48W, 3.68S), and San Antonio, Loreto (74.38W, 3.68S). Both are almost as hot as Puerto Curaray. It is very humid east of Nauta, Loreto, near the river Yavarı´ Mirim close to the border to Brazil (73.18W, 4.68S). Lima, the capital of Peru, is located at the Pacific coast in a moderate climatic zone with mean
Climatic Data for Suriname
City
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Paramaribo (capital)
26.1
26.3
14
28.22
83.3
66.4
Totness
26.1
26.3
14
28.08
83.0
66.1
DOI 10.1002/jps
957
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75
2 5 13 2 6 13
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
958
ZAHN ET AL.
Figure 15. Daily partial vapour pressure fluctuations—Capata´rida, Venezuela.
Figure 14. Daily temperature fluctuations—Puerto Piritu, Venezuela.
temperatures fluctuating between 118C and 238C, and mean PD values below 10 hPa from June to October. Details are listed in Table 9.
equal high levels throughout the country but more pronounced towards the Atlantic coast (see Tab. 10).
Suriname
Venezuela
The dominating climate is in Group Am. The hottest place is southeast of Paramaribo, the capital of Suriname, at 55.18W, 5.68N, with minimum mean temperatures of less than 248C during the nights in January to March, and maximum mean temperatures of more than 308C during the days in September and October (see Fig. 13). Partial vapour pressures are at
The northern half of Venzuela belongs to Group BS, the coast to the Caribbean Sea in the north to Group BW, the central plains and the Guiana Highlands in the southeast to Group Aw. One of the extreme places is at the southwest corner of the country at 66.48W, 1.18N, southeast of El Carmen (Amazonas State), near the borders to Brazil and Colombia, with the highest mean
Table 11. Climatic Data for Venezuela
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
Capata´rida, Falco´n
27.3
27.4
10
29.24
80.7
68.8
Caracas (capital)
26.3
26.6
13
24.96
73.0
58.8
El Carmen, Amazonas
25.2
25.4
18
28.60
89.3
67.3
El Sama´n, Apure
27.0
27.4
9
26.82
75.0
63.1
Paraguaipoa, Zulia
27.3
27.6
9
27.61
76.2
65.0
Puerto Pirito, Anzoa´tegui
27.3
27.7
8
25.13
69.0
59.2
Santa Marı´a de Iripe, Gua´rico
27.2
27.6
9
24.75
68.6
58.3
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
RH [%]
RH [%] at 308C
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
6 2 9 11 19 28 4 4 11 3 11 19 0 8 15 10 18 27 12 20 29
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
Table 12.
959
Climatic Data for China
City, Provinces
T [8C]
MKT [8C]
Hong Kong (HKG)
23.4
24.4
Macau (MAC)
23.0
24.1
Mengla, Yunnan
21.1
22.1
Nanchang, Jiangxi
18.0
21.5
Nanchong, Sichuan
18.1
20.9
Sanya, Hainan
26.0
26.4
Shanghai, Shanghai
15.8
19.0
YT [%] T ¼ 258C T ¼ 308C 3 23 4 24 13 36 16 40 19 43 5 14 32 58
PD [hPa]
RH [%]
22.21
77.4
22.12
78.9
19.34
77.3
15.73
76.0
15.29
73.7
27.16
80.7
13.96
77.8
RH [%] at 258C at 308C
Testing Condition [8C/% RH]
70.1 52.1 69.8 52.1 61.0 45.5 49.6 37.0 48.2 36.0 85.7 64.0 44.0 32.9
25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65
YPD [%] 14 25 14 25 2 43 21 76 24 81 30 2 36 98
relative humidity due to high partial vapour pressures at lower temperatures. A hot climate has been found near El Sama´n (Apure State), at 68.68W, 7.88N and also at 65.28W, 8.98N, northeast of Santa Marı´a de Iripe (Gua´rico State). The climate south of Paraguaipoa (Zulia State), at 71.9 W, 11.2 N, at the coast to the Golfo de Venezuela, in the northwest corner of the country, is also of interest. Caracas, the capital of Venezuela, northeast of 67.58W, 10.08N, at the coast to the Caribbean Sea, presents a climate which is less humid than the extreme places identified. The hottest place is in the area of 65.28W, 10.08N, which is west of Puerto Piritu (Anzoa´tegui State) at the coast to the Caribbean Sea. There is not one single day when the mean maximum temperature is below 308C, while mean minimum temperatures never decrease to less than 248C (during the night in January and February) (see Fig. 14).
The place with the highest mean partial vapour pressures in Venezuela is at 70.88W, 11.28N, west of Capata´rida (Falco´n State), in the northwest part of the country at the coast to the Golfo de Venezuela. Mean maximum water vapour pressures reach almost 32 hPa in September and October, and mean minimum values can be found above 26 hPa in January and February (see Fig. 15). The climatic parameters for Venezuela are listed in Table 11.
Figure 16. Daily temperature fluctuations—Sanya, Hainan, China.
Figure 17. Daily partial vapour pressure fluctuations—Sanya, Hainan, China.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
CHINA Mainland China is extremely diverse from a climatic point of view: from tropical parts in the south (Group Am) to regions with severe dry winters (Group Dw) in the northeast, and humid subtropical regions (Group Cfa) in between.
960
ZAHN ET AL.
Figure 19. Daily partial vapour pressure fluctuations—Cape Saint Lucia, Natal, South Africa.
Figure 18. Daily temperature fluctuations—Cape Saint Lucia, Natal, South Africa.
Shanghai at the east coast, Macau and Hong Kong in the south east and the city located in the most southern corner of the mainland, Mengla close to the border to Laos, have been selected. Another city in the south is Sanya on the southern coast of the island Hainan. In addition to these cities, two cities identified as hot and humid ‘spots’ in the centre of the land mass have been analysed, Nanchang in Jiangxi, and Nanchong in Sichuan. The results of the calculations are presented in Table 12 for both testing at 258C/60% RH and 308C/ 65% RH. As can be concluded from the table, the most loading place in China is Sanya on the island Hainan in the most southern part of the country.
Mean maximum temperatures reach more than 308C in four consecutive months (May to August), and mean minimum temperatures never drop below 208C (in January, see Fig. 16). The southern part of this island belongs to Ko¨ppen Group Am, the northern part to Group Aw. The maximum PD values increase to more than 33.0 hPa during 3 months (June to August, see Fig. 17). To calculate the adequate relative humidity for long-term stability testing, the mean partial water vapour pressure calculated for Sanya (27.16 hPa) is used at the standard testing temperature 308C to get 64.0% RH. Testing at 308C would include a safety margin of 14% added to the MKT, and
Table 13. Climatic Data for South Africa
T [8C]
MKT [8C]
Cape Saint Lucia, KwaZulu-Natal Cape Town, Western Cape
22.3
22.6
16.7
17.0
De Aar, Northern Cape
16.7
19.2
Durban, KwaZulu-Natal
20.6
21.0
Kimberley, Northern Cape
17.6
20.3
Little Namaland, Northern Cape
21.4
23.6
Musina (Messina), Limpopo
21.5
22.8
Port Elizabeth, Eastern Cape
18.8
19.1
Pretoria, Gauteng
17.5
19.2
City, Provinces
YT [%] T ¼ 258C T ¼ 308C
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
11 33 47 77 30 57 19 43 23 48 6 27 10 32 31 57 30 56
PD [hPa]
RH [%]
20.66
76.9
14.96
78.8
8.26
43.5
19.28
79.4
9.76
48.4
9.65
37.8
15.73
61.3
16.27
75.1
10.70
53.4
RH [%] at 258C at 308C
Testing Condition [8C/% RH]
YPD [%]
65.2 48.6 47.2 35.2 26.1 19.4 60.8 45.4 30.8 23.0 30.4 22.7 49.6 37.0 51.3 38.3 33.8 25.2
25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65
8 34 27 85 130 234 1 43 95 183 97 186 21 76 17 70 78 158
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
961
Figure 21. Daily partial vapour pressure fluctuations—Mbandaka, Dem. Rep. Congo.
Figure 20. Daily temperature fluctuations—Dar es Salaam, Tanzania.
SOUTH AFRICA testing at 65% RH would include a safety margin of 2% for PD. Sanya, however, presents an extreme climate compared to the other parts of the country.
Table 14.
Climatic Data for SADC (Without South Africa)
Country: City
T [8C]
MKT [8C]
Angola: Luanda
25.1
25.7
Botswana: Gaborone
19.5
21.5
Dem Rep Congo: Boma
24.6
24.8
Kinshasa
24.1
24.4
Mbandaka
24.5
24.8
Lesotho: Maseru
13.8
15.9
Malawi: Lilongwe
21.9
22.7
Mauritius
24.5
24.7
Mozambique: Maputo
22.8
23.6
Namibia: Windhoek
21.5
23.3
Swaziland: Mbabane
18.9
20.1
Tanzania: Dar es Salaam
26.2
26.6
Zambia: Lusaka
21.7
22.7
Zimbabwe: Harare
19.2
20.2
DOI 10.1002/jps
There is only a small part of South Africa presenting a Group Cfa climate, that is the Durban area near the Indian Ocean. The most humid place identified in South Africa is Cape
YT [%] T ¼ 258C T ¼ 308C 3 17 16 40 1 21 3 23 1 21 58 89 10 32 1 22 6 27 7 29 24 49 6 13 10 32 24 48
PD [hPa]
RH [%]
23.48
73.6
11.84
52.2
26.25
84.9
24.14
80.6
26.91
87.5
8.01
50.9
16.92
64.4
23.39
76.2
19.22
69.1
7.27
28.4
16.01
73.1
25.26
74.2
15.27
58.9
13.63
61.4
RH [%] at 258C at 308C
Testing Condition [8C/% RH]
74.1 55.3 37.4 27.9
25/60 30/65 25/60 30/65
19 18 61 133
82.8 61.8 76.2 56.8 84.9 63.4 25.3 18.9 53.4 39.8 73.8 55.1 60.6 45.3 22.9 17.1 50.5 37.7 79.7 59.5 48.2 36.0 43.0 32.1
25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65
28 5 21 14 29 3 137 245 12 63 19 18 1 44 162 280 19 72 25 9 25 81 40 103
YPD [%]
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
962
ZAHN ET AL.
Table 15. Climatic Data for Cuba
City
T [8C]
MKT [8C]
YT [%] T ¼ 308 C
PD [hPa]
RH [%]
RH [%] at 308C
Testing Condition [8C/% RH]
YPD [%]
Camagu¨ey Santiago de Cuba West of Santa Clara
26.1 26.6 25.3
26.4 26.9 25.7
14 12 17
24.69 25.98 23.40
72.9 74.6 72.6
58.1 61.2 55.1
30/65 30/65 30/65
12 6 18
Saint Lucia at the west coast north of Durban. Mean maximum temperatures can go up to 278C at noon in February, combined with a mean maximum partial vapour pressure of almost 26 hPa. Mean minimum temperatures decrease just below 188C at night in July (see Figs. 18 and 19). In the centre of South Africa, the Group BS is dominating. Pretoria and Johannesburg are located there, while Cape Town is located in a Group Bsk climate. At the south coast, South Africa presents a Group Cfb climate (see Tab. 13).
SADC Tropical Climatic Group Af is to be found north and south of the equator, surrounded by Group Aw regions. The hottest place in SADC identified so far is Dar es Salaam (Group Aw) in Tanzania (see Fig. 20). Mean maximum temperatures increase to 31.68C in February, and never decrease below 218C. The most appropriate temperature for longterm stability testing of medicinal products to be marketed in Tanzania and SADC is 308C. That value includes a safety margin of 13% added to the highest MKT calculated for Dar es Salaam. The most humid place is Mbandaka (Group Af): Values for partial water vapour pressure increase
to almost 29.0 hPa in May, and decrease to a minimum of 24.9 hPa in July (see Fig. 21). The mean PD value of 26.91 hPa calculated for Mbandaka at the testing temperature of 308C would result in a relative humidity of 63.4% RH. Testing at 308C/65% RH would include a safety margin of 3% for PD (see Tab. 14).
CARIBBEAN ISLANDS All of the northwestern Caribbean Islands are in Ko¨ppen Group Aw, the southeastern islands belong to Group Af. Cuba Three representative places have been selected, one southeast of Havanna, west of Santa Clara, another in the centre of the eastern part near Camagu¨ey and the third near Santiago de Cuba at the southeast coast of the island. The island belongs to Group Aw. For long-term stability testing of medicinal products to be marketed in Cuba, 308C is appropriate. This value includes a safety margin of 12% added to the highest MKT calculated for Cuba (see Tab. 15). The mean PD value of 25.98 hPa calculated for Santiago de Cuba at the testing temperature of
Figure 22. Daily temperature fluctuations—Curac¸ao, The Netherlands Antilles.
Figure 23. Daily partial vapour pressure fluctuations—Curac¸ao, The Netherlands Antilles.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
Table 16.
963
Climatic Data for Curac¸ao and Puerto Rico
Island
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Curac¸ao
27.5
27.6
9
29.46
80.2
69.4
Puerto Rico
26.8
26.9
12
27.51
78.2
64.8
308C would result in a relative humidity of 61.2% RH. Testing at 308C/65% RH would include a safety margin of 6% for PD (see Tab. 15). Curac¸ao The island’s climate is similar but more extreme than the other Caribbean Islands. Mean temperatures can go up to 29.88C in October, and never decrease to less than 25.48C in February (see Fig. 22). Mean partial pressures reach a maximum of 32.27 hPa in September (see Fig. 23 and Tab. 16). Puerto Rico The climate is characterised by high temperatures throughout the year, that is, fluctuation between 24.38C in February and 29.48C in September, and high humidity, that is, up to 31.14 hPa. It belongs to Ko¨ppen Group Aw. Testing at 308C/65% RH would include a safety margin of 12% for the MKT, but 0% for PD (see Tab. 16).
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75
6 1 8 0 8 16
DISCUSSION As a first step to develop an adequate long-term stability testing condition for hot and humid countries and regions, meteorological data based on monthly mean values of temperatures and dewpoints led to the evaluation of a climate of a particular country or region. The next step was the identification of the hottest and/or most humid place in a country assuming that a medicinal product can be shipped and stored everywhere within the boundaries of a country where the product is allowed to be marketed. This concept can be extended in principle to a group of associated countries, for example ASEAN. As the hottest and most humid place is representing the upper end of potential climatic conditions instead of an average of all climates, there will be a difference between the testing and the climatic conditions for places with lower humidities. This difference can be calculated by applying the safety margin Y for temperature and partial water vapour pressure PD. Assessing the results of these calculations facilitates the assignment of a country to either Climatic Zone IVA or IVB.
Table 17. Long-Term Testing Conditions for ASEAN Member Countries
Country Brunei Cambodia Indonesia Laos Malaysia Myanmar Philippines Singapore Thailand Vietnam
DOI 10.1002/jps
308C/65% RH, 308C/70% 308C/75% RH, CZ IVA RH CZ IVB þ þ þ þ
Country þ
þ
þ
Table 18. Long-Term Testing Conditions for South American Countries
þ þ þ
Bolivia Brazil Colombia Ecuador Guyana Peru Suriname Venezuela
308C/65% RH, 308C/75% RH, CZ IVA 308C/70% RH CZ IVB þ þ
þ þ
þ þ þ þ
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
964
ZAHN ET AL.
Table 19. Long-Term Testing Conditions for China Country
Country
þ þ þ
CONCLUSION As a result of this risk assessment, an appropriate long-term stability testing condition is proposed for each selected country (see Tabs. 17–21). In addition to the long-term testing condition 258C/ 60% RH for markets in Climatic Zone II, the hot and humid countries are categorized as either Table 20. Long-Term Testing Conditions for SADC Member Countries 258C/60% RH, CZ II þ þ þ þ þ þ þ
308C/65% RH, CZ IVA
Cuba Curac¸ao Puerto Rico
An adequate long-term testing condition should be established by balancing the need to provide the patients with stable medicinal products with the need to reduce costs of goods for industry. If the testing conditions were to be too stressful, the consequence would be unnecessary protective packaging (e.g. double aluminium blister), a reduced shelf-life (e.g. 2 years instead of 3 years) or too restrictive storage recommendations (store below 258C). Careful analysis of the climate in all tropical and subtropical parts of the world, and calculation of the relative humidity at the standard testing temperature 308C offer the possibility to evaluate the ‘‘probability of failure’’ of a particular product in a particular market, that is the risk for insufficient stability.
Angola Botswana Dem. Rep. Congo Lesotho Malawi Mauritius Mozambique Namibia South Africa Swaziland Tanzania Zambia Zimbabwe
Long-Term Testing Conditions for Caribbean
308C/65% RH, CZ IVA
China (Mainland) Hong Kong Macau
Country
Table 21. Islands
308C/65% RH, CZ IVA þ þ þ þ þ þ
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
þ
308C/70% RH þ þ
belonging to category CZ IVA, for which testing at 308C/65% RH would be sufficient, or to CZ IVB, if testing at 308C/75% RH were required. The long-term testing condition 308C/70% RH has been kept in these tables to facilitate the risk assessment for those products, for which 308C/75% RH data are not yet available. This would provide manufacturers with more flexibility in the transition period when testing at 308C/70% RH could eventually be replaced by the condition 308C/75% RH.
REFERENCES 1. WHO. 1996. Guidelines for stability testing of pharmaceutical products containing well established drug substances in conventional dosage forms. WHO Technical Report Series 863, Annex 5. 2. WHO. 2003. Report of the 37th meeting of the WHO Expert Committee on Specifications for Pharmaceutical Preparations, Geneva, 22–26 October 2001. WHO Technical Report Series 908, page 13. 3. ICH Q1F. 2003. Stability data package for registration applications in Climatic Zones III and IV. http://www.ich.org/LOB/media/MEDIA416.pdf 4. Slamet LS, et al. 2004. Stability testing for hot and humid climates. WHO Drug Inf 18:113– 116. 5. WHO. A WHO consultation of stability studies in a global environment was held in Geneva on 13–14 December 2004. 6. Resolution RE No. 1, dated 29 July 2005, published in the Official Gazette of the Union, Supplement to No. 146- Section 1, Brası´lia- DF, on 1 August 2005. 7. WHO. 40th WHO Expert Committee on Specifications for Pharmaceutical Preparations, Geneva, 24–28 October 2005. ¨ ber eine fu¨r die Haltbarkeit 8. Schumacher P. 1972. U von Arzneimitteln maßgebliche Klimaeinteilung. Pharm Ind 34:481–483. 9. Grimm W. 1986. Storage conditions for stability testing (Part 2). Drugs made in Germany 29:39–47. 10. Grimm W. 1998. Extension of the International Conference on Harmonization Tripartite Guideline DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
965
for stability testing of new drug substances and products to countries of Climatic Zones III and IV. Drug Dev Ind Pharm 24:313–325. 11. Ref 1, Table 2 and Table 3, page 70. 12. The United States Pharmacopeia (USP) 28 (2005), General Chapter 1150 Pharmaceutical Stability. 13. Uppala SM, et al. 2005. The ERA-40 Reanalysis. Q J R Meteorological Soc 131:2961–3012.
14. Hardy R. 1998. ITS-90 Formulations for vapor pressure, frostpoint temperature, dewpoint temperature, and enhancement factors. In Proceedings of the Third International Symposium on Humidity & Moisture, Teddington, London. 15. Haynes JD. 1971. Worldwide virtual temperatures for product stability testing. J Pharm Sci 60:927– 929.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
AstraZeneca R&D So¨derta¨lje, Regulatory CMC, So¨derta¨lje, Sweden
2
European Centre for Medium-Range Weather Forecasts, Reading RG2 9AX, United Kingdom
3
AstraZeneca R&D So¨derta¨lje, So¨derta¨lje, Sweden
Received 19 April 2005; revised 24 November 2005; accepted 21 December 2005 Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20584
ABSTRACT: The external stability risk factors heat and moisture are evaluated with respect to the development of pharmaceutical products intended to be marketed in tropical and subtropical countries. The mean daily temperatures and dewpoints measured four times per day at selected places in Southeast Asia, South America, China, Southern Africa and the Caribbean are used to calculate the daily and monthly fluctuations of temperature and partial water vapour pressure, the mean kinetic temperature and the relative humidity. Based on these data, the hottest and the most humid place in each country or region are identified to reflect the worst case for the specific region. A formula to calculate safety margins for temperature and partial vapour pressure is introduced taking into consideration the difference between measured meteorological parameters and the stability testing conditions. An appropriate long-term stability testing condition is proposed for each selected country, related to the worst case for each specific region and the safety margins, as well as its classification in either Climatic Zone IVA or IVB. ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 95:946–965, 2006
Keywords: physicochemical properties; tropical climate; hot and humid; high humidity; testing condition; safety margin; climatic zone; thermodynamics; stability; regulatory science
INTRODUCTION In more than 30 years of ongoing discussions, pharmaceutical manufacturers could not agree with regulators on a common long-term stability All web addresses in References have been accessed in March 2006. Gijsbert M. Slappendel is a Pharmacy student placed for internship at Regulatory CMC during December 2004 to February 2005 and Harold M. Smeenge is a Pharmacy student placed for internship in 2004 at Regulatory Affairs, So¨derta¨lje, Sweden. Correspondence to: Manuel Zahn (Telephone: þ46-8-552 57640; Fax: þ46-8-552 54660; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 95, 946–965 (2006) ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association
946
testing condition, which would be adequate for hot and humid countries. While testing at 308C/70% RH became common practice in industry, and was also recommended by the World Health Organization (WHO)1 in 1996, in 2001 the WHO changed the recommended testing condition for hot and humid countries to 308C/65% RH2. Subsequently, the International Conference on Harmonisation (ICH) Stability Guideline Q1F recommended the same condition3. Regulators representing the Association of South East Asian Nations (ASEAN), however, decided in 2004 not to follow this advice, and required in their published guideline4 to test at 308C/75% RH instead. In December 2004, the WHO encouraged5 manufacturers and regulators
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
Table 1.
Criteria used to Classify Climatic Zones and Testing Conditions
CZ
Definition
I II
Temperate climate Subtropical and Mediterranean climate Hot and dry climate Hot and humid climate
III IV
947
Criteria, Mean Annual Temperature Measured in the Open Air/Mean Annual Partial Water Vapour Pressure
Testing Conditions [8C/% RH]
158C/11 hPa >15 to 228C/>11 to 18 hPa
21/45 25/60
>228C/15 hPa >228C/>15 hPa
30/35 30/70
to propose a solution, which could be acceptable to all parties involved. In August 2005, the Brazilian authorities introduced 308C/75% RH as the longterm condition required for Brazil6. The 40th WHO Expert Committee on Specifications for Pharmaceutical Preparations met in Geneva in October 20057, and decided to split the current Climatic Zone IV (hot and humid countries) into CZ IVA, for which 308C/65% RH will remain the standard long-term testing condition, and CZ IVB, for which, if justified, 308C/75% RH will become the testing condition. In the following, the climate of a selected number of hot and humid countries in Climatic Zone IV are presented and analysed, and the most appropriate testing conditions for long-term stability studies are proposed to facilitate the categorisation of different hot and humid countries in Climatic Zone IVA and IVB.
CLIMATIC ZONES
19728 and Wolfgang Grimm in 19869 and in 199810 (see Tab. 1) when they defined four different long-term test conditions, which match with the climatic conditions of the target markets categorised in just four different climatic zones. This concept is described in regulatory guidelines11 and pharmacopoeias12 and became an established standard in developing pharmaceutical products. As a result of the above-mentioned WHO proposal to split the Climatic Zone IV into two different zones, IVA and IVB, there is a need for criteria to categorize a country to one of these CZs. We suggest to apply the criteria listed in Table 2. Ko¨ppen It is proposed to use the Ko¨ppen classification of world climates in order to be able to distinguish between different regional climates in Climatic Zone IV countries as being either ‘‘Moist tropical’’ (Group Af, Am and Aw) or ‘‘Humid subtropical’’ (Cfa).
Schumacher/Grimm In order to be able to reduce the amount of stability testing, the number of different testing conditions must be reduced to a sufficient extent. This has been proposed by Paul Schumacher in Table 2.
DOI 10.1002/jps
CALCULATION OF SAFETY MARGINS For temperature, the following equation can be applied to calculate a safety margin:
Proposal for Criteria to Split Climatic Zone IV
CZ
Definition
Criteria, Mean Annual Temperature Measured in The Open Air/Mean Annual Partial Water Vapour Pressure
IVA IVB
Hot and humid climate Hot and very humid climate
>228C/27 hPa >228C/>15 to 27 hPa
Testing Conditions [8C/% RH] 30/65 30/75
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
948
ZAHN ET AL.
YT ¼ ðTS TÞ 100=T YT, margin of safety for temperature [%]; TS, stability testing storage temperature; T, temperature measured in the environment or calculated as MKT. YT > 0
if
TS > T
YT ¼ 0
if
TS ¼ T
There is a probability of failure if YT < 0. The same principle can be applied to other parameters like partial water vapour pressure. The amount of safety margin required is dependent on: the impact of the environment on the product during shelf-life (variability in loading), and on the other hand on the strength of a particular batch of a product to resist heat and moisture (‘‘variability in resistance’’). The distributions of the loading and the resistance result in the ‘‘probability of failure’’ of a particular product in a particular market. The variability in resistance can be neglected, as the batch-to-batch variability of a product’s stability is too low to be taken into consideration. As a consequence, the focus is on the variability in loading, for example, the fluctuations of heat and moisture in the environment or the frequency of extreme temperatures and humidities during the shelf life of a particular product.
CLIMATIC DATA
Figure 1. Daily temperature fluctuations—El Nido, The Philippines.
means at 00UTC, 06UTC, 12UTC and 18UTC. ‘Screen level’ is normally about 2 m above ground where the standard meteorological instruments are placed. Vapour pressures—applying an updated version of Wexler’s equation14—and relative humidities were then determined from analysed temperatures and dewpoints using basic thermodynamic relationships. The additional parameter mean kinetic temperature (MKT), which includes the reaction rate constants in the evaluation of heat effects on pharmaceutical products, was calculated by applying an equation derived by Haynes15 based on the Arrhenius equation. The activation energy Ea ¼ 83.144 kJ/ mol has been used. The climatic parameters were then analysed in order to identify: the daily and monthly fluctuations of temperature and vapour pressure; their mean maximum values; the place representing the most loading climatic condition in each country or region.
Quantification of the variability in loading starts with the determination of key meteorological parameters at different times of the day and the year, leading to identification of the most loading part of a country or region. Global climatic data have been extracted from the ERA-40 reanalysis of ECMWF.13 Reanalysis is a process in which global weather observations, near the ground and in the free atmosphere, are assembled to form a regular mesh covering the earth with a resolution of about 125 km and at 60 vertical levels using a sophisticated meteorological forecasting system. Measurements take place every 6th hour. Reanalysed screen level temperatures and dewpoints from 1979 to 2002 were averaged into monthly
Figure 2. Daily partial vapour pressure fluctuations—El Nido, The Philippines.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
949
Table 3. Climatic Data for ASEAN Member Countries
Country
City
Brunei
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
25.5
25.8
16
27.35
83.8
64.4
Cambodia
Phnom Penh
26.8
27.2
11
27.69
78.5
65.2
Indonesia
Jakarta, Java
27.1
27.2
10
28.96
80.6
68.2
Kolbano, Timor
27.6
27.7
8
27.49
74.5
64.7
SW Sulawesi
27.4
27.5
9
28.82
78.9
67.9
Palembang, Sumatra
26.7
26.9
11
29.17
83.2
68.7
Surabaya, Java
27.6
27.7
8
29.07
78.7
68.4
Laos
Vianchan (Vientiane)
24.9
25.6
17
24.22
77.0
57.0
Malaysia
Kuala Lumpur
26.1
26.5
13
27.91
82.7
65.7
Myanmar
Yangon
26.6
27.5
9
25.66
73.5
60.4
Philippines
Cebu, Cebu
27.3
27.4
9
30.18
83.1
71.1
Davao, Mindanao
25.7
25.9
16
28.64
86.7
67.4
El Nido, Palawan
28.0
28.1
7
30.28
80.2
71.3
Manila, Luzon
26.9
27.0
11
29.33
82.9
69.1
Roxas, Panay
27.4
27.6
9
29.10
79.7
68.5
27.2
27.5
9
29.14
80.7
68.6
Singapore
Thailand
Bangkok
27.9
28.4
6
27.17
72.3
64.0
Vietnam
Hanoi
23.8
24.9
21
24.08
81.9
56.7
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70
1 9 17 0 7 15 5 3 10 0 8 16 4 3 11 5 2 9 5 2 10 14 23 32 1 7 14 8 16 24 9 2 6 4 4 11 9 2 5 6 1 9 5 2 10 5 2 9 2 9 17 15 24 (Continued)
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
950
ZAHN ET AL.
Table 3. (Continued)
City
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Ho Chi Minh City (Saigon)
27.3
27.6
9
28.11
77.6
66.2
Country
Testing Condition [8C/% RH]
YPD [%]
30/75 30/65 30/70 30/75
32 2 6 13
T, mean temperature, calculated by using the sum of 48 measured temperatures (4 temperatures per day for each month), divided by 48; MKT, mean kinetic temperature, calculated as described above; YT, safety margin for temperature, calculated using the MKT versus the testing temperature 308C (for details please refer to chapter Calculation of safety margins); PD, mean partial water vapour pressure, calculated by taking the dewpoints; RH, mean relative humidity, calculated by using the saturation vapour pressure PS at the measured temperature, and the value for PD found in the previous column; RH at 308C, mean relative humidity, calculated by using the saturation vapour pressure PS at the testing temperature 308C, and the value for PD found in the previous column; YPD, safety margin for partial vapour pressure, calculated using the meteorological PD value versus the PD value calculated for the respective testing condition found in the previous column.
SELECTION OF COUNTRIES For the purpose of this exercise, the following countries with tropical and subtropical moist climates have been selected for detailed evaluation, that is countries in Group A and Cfa according to Ko¨ppen:
All of the 10 ASEAN member countries, Brazil and other countries in South America, China (Mainland), Hong Kong and Macau, South Africa, and most of the other SADC member countries in southern Africa, Some Caribbean Islands.
the southeast that belong to Group Aw. Kolbano (on the island Timor in the south-west, west of Kupang) is the place with the highest monthly mean temperatures in two consecutive months (October and November) reaching a peak at 31.58C. The highest monthly mean values for PD in Indonesia have been found at the same place: 31.51 hPa in January with values above 31 hPa in four consecutive months (December to March). The Philippines
All of the Indonesian islands belong to the Ko¨ppen Group Af with the exception of some islands in
The Philippines can be characterised as tropical islands with only minor fluctuations of temperatures and partial water vapour pressure during the day or year. Mean maximum temperatures normally do not exceed 318C, and never drop below 238C. PD values can be found in the range of 25.0 to 33.0 hPa. All of the islands belong to Ko¨ppen Group Am; just a small part of the northern island Luzon is characterised by Group Aw.
Figure 3. Daily Brazil.
fluctuations—Pau,
Figure 4. Daily partial vapour pressure fluctuations—llha Macuapanim, Brazil.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
ASEAN Indonesia
temperature
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
951
Table 4. Climatic Data for Brazil
City, Federal State
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Bele´m, Para´
26.7
26.8
12
29.29
83.4
69.0
Brası´lia (capital)
22.9
23.4
28
18.69
67.1
44.0
Fortaleza, Ceara´
27.1
27.1
11
28.49
79.5
67.1
Ilha de Marajo´, Para´
26.5
26.7
13
29.97
86.6
70.6
Ilha Macuapanim, Amazonas
26.0
26.3
14
30.24
89.7
71.2
Macapa´, Amapa´
26.4
26.7
12
29.24
84.9
68.8
Manaus, Amazonas
27.0
27.4
10
28.33
79.2
66.7
Natal, Rio Grande do Norte
26.7
26.8
12
28.00
79.7
65.9
Pau, Rio Grande do Norte
28.3
28.9
4
22.02
57.1
51.8
Poˆrto Alegre, Rio Grande do Sul
19.7
20.4
47
18.33
79.8
43.2
Recife, Pernambuco
25.9
26.1
15
25.56
76.3
60.2
Rio de Janeiro, Rio de Janeiro
22.9
23.3
29
22.41
80.1
52.8
Salvador, Bahia
26.0
26.1
15
26.60
79.1
62.6
Sa˜o Joa˜o, Amapa´
26.8
26.8
12
28.82
81.7
67.9
Sa˜o Luı´s, Maranha˜o
27.0
27.0
11
29.06
81.5
68.4
Sa˜o Marcelino, Amazonas
25.3
25.5
18
29.06
90.1
68.4
Sa˜o Paulo, Sa˜o Paulo
20.6
21.1
42
18.70
77.2
44.0
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65
6 2 9 48 59 70 3 4 12 8 1 6 9
30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65
2 5 6 2 9 3 5 12 1
30/70 30/75 30/65 30/70 30/75 30/65
6 14 25 35 45 51
30/70 30/75 30/65 30/70 30/75 30/65
62 74 8 16 25 23
30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65
33 42 4 12 20 4 3 11 5 2 10 5 2 10 48 (Continued)
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
952
ZAHN ET AL.
Table 4. (Continued)
City, Federal State
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Teresina, Piauı´
27.2
27.6
9
25.77
71.5
60.7
Uraricoera, Roreima
27.1
27.5
9
23.72
66.3
55.9
The most loading place in the Philippines and probably in the whole ASEAN region identified so far is El Nido in the north of the island Palawan with mean temperatures between 25.98C and 31.08C (average per year: 28.08C), and PD values between 27.3 and 33.1 hPa (average per year: 30.3 hPa). As an example of the graphical presentation of climatic parameters, the daily temperatures and partial vapour pressures per month for El Nido (mean values 1979–2001) are shown in Figures 1 and 2.
Testing Condition [8C/% RH]
YPD [%]
30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
59 70 7 15 24 16 25 34
A list of the key climatic parameters measured and calculated for all ASEAN member countries (Tab. 3) facilitates the selection of the most loading place.
calculated at the place with the highest temperature (El Nido, Philippines: 28.18C), or a safety margin of 6% to the maximum MKT found (Bangkok, Thailand: 28.48C), to get a temperature for long-term stability testing of 308C. To calculate an adequate relative humidity for stability testing, the highest mean value for PD (El Nido, Philippines: 30.28 hPa) is kept constant when moving to a testing temperature of 308C, at which the saturation water vapour pressure is 42.47 hPa, which results in a relative humidity of 71.3% (RH ¼ PD 100/PS). Testing at 308C/70% RH would represent the climatic conditions at the most loading place in the Philippines and the ASEAN region. Seventy percent of relative humidity, however, would give a negative safety margin of 2%.
Testing Conditions for ASEAN
BRAZIL
Comparison of Places in ASEAN
With regard to an appropriate testing temperature, a safety margin of 7% is added to the MKT
Near the equator, Brazil belongs to Ko¨ppen Group Af and Am, north and south of the equator to Group Aw, in the south-east to Group Cfa and in between to Group Cw.
Figure 5. Daily temperature fluctuations—Fortin Paredes, Bolivia.
Figure 6. Daily partial vapour pressure fluctuations—Northeast corner, Bolivia.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
953
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
Table 5. Climatic Data for Bolivia
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Cavinas, Beni
25.3
25.7
17
26.57
82.5
62.6
Fortı´n Paredes, Santa Cruz
25.5
26.2
14
21.21
64.9
49.9
Mapiri River, Pando
25.4
25.8
16
27.35
84.0
64.4
Northeast corner, Pando
25.4
25.8
16
27.46
84.6
64.7
The highest temperatures in Brazil have been identified in Pau, Rio Grande do Norte, south of Fortaleza. Four consecutive months (from September to December) show mean maximum temperatures above 34.08C, and no mean minimum temperature below 23.78C is found (see Fig. 3). As a consequence, the adequate temperature for long-term stability testing of medicinal products to be marketed in Brazil would be 308C. That value includes a safety margin of 4% added to the highest MKT calculated for Brazil. The highest values for PD are to be found near the Amazonas river, in particular on the Amazonas island Ilha Macuapanim west of Jau´ National Park, where the PD values never decrease below 27.7 hPa and can go up to 32.5 hPa with a mean of 30.2 hPa (see Fig. 4). This island, however, is not populated in contrast to the other extreme places identified like Bele´m, the capital of Para´, Fortaleza, the capital of Ceara´,
Figure 7. Colombia. DOI 10.1002/jps
Daily temperature fluctuations—Arjona,
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
4 12 20 30 40 50 1 9 17 1 8 16
Macapa´, the capital of Amapa´, Manaus, the capital of Amazonas, Natal, the capital of Rio Grande do Norte and Sa˜o Luı´s, the capital of Maranha˜o.
All of these major cities, however, show a small but positive safety margin for the testing condition 308C/70% RH. In the following, the analysis for all major cities in Brazil, all capitals of federal states in tropical climates, as well as the hottest and most humid places are presented. The safety margin for PD is calculated for testing conditions 308C/65% RH (first line), 308C/70% RH (second line) and 308C/ 75% RH (third line) for each selected place (see Tab. 4).
OTHER COUNTRIES IN SOUTH AMERICA Bolivia The Andes mountains in the southwest of Bolivia are creating a highland climate (Ko¨ppen Group H), while the lowland plains of the Amazon Basin in the northeast are responsible for a humid
Figure 8. Daily partial vapour pressure fluctuations—San Juan, Antioquia, Colombia. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
954
ZAHN ET AL.
Table 6. Climatic Data for Colombia
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Arjona, Bolı´var
27.4
27.5
9
29.87
81.8
70.3
Bogota´ (capital)
19.1
19.4
55
17.50
79.1
41.2
El Banco, Magdalena
26.6
26.8
12
28.84
82.9
67.9
La Pedrera, Amazonas
25.4
25.6
17
28.81
89.0
67.8
Meta, west part
26.0
26.3
14
26.81
79.7
63.1
Sahagu´n, Co´rdoba
26.9
27.1
11
29.53
83.3
69.5
Santa Clara, Amazonas
25.5
25.7
17
28.97
89.0
68.2
San Juan, Antioquia
27.2
27.3
10
29.96
82.9
70.5
Santa Marta, Magdalena
26.7
26.8
12
28.92
82.3
68.1
Sa. de Chiribiquete, Guaviare
25.7
26.0
15
24.86
75.2
58.5
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
8 1 7 58 70 82 4 3 10 4 3 11 3 11 19 7 1 8 5 3 10 8 1 6 5 3 10 11 20 28
tropical wet and dry (savanna) climate (Group Aw). In between, there is a moist climate with mild dry winters (Group C). The hottest place is close to the border to Paraguay in the southeast, east of Fortı´n Paredes,
department Santa Cruz (59.68W, 19.28S) (see Fig. 5). The most humid place identified is in the northeast corner of the country, close to the border to Brazil, in the department Pando (65.38W,
Figure 9. Daily temperature fluctuations—Sarayacu, Pastaza, Ecuador.
Figure 10. Daily partial vapour pressure fluctuations—Yasuni N.P., Orellana, Ecuador.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
955
Table 7. Climatic Data for Ecuador
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Jipijapa, Manabı´
23.7
23.8
26
25.21
86.0
59.4
Sarayacu, Pastaza
26.5
27.0
11
23.69
68.4
55.8
Quito (capital)
18.7
19.1
57
17.25
80.1
40.6
Yasunı´ N.P.
25.2
25.5
18
27.40
85.7
64.5
10.28S). In 6 months from November to April, the mean maximum values for partial pressure exceed 30 hPa (see Fig. 6). West of this corner is another humid place, east of the river Mapiri, south of the border to Brasil (66.48W, 10.28S). The other extreme place assessed is east of Cavinas, department Beni (66.48W, 12.58S) (see Tab. 5).
Colombia Colombia presents a highland climate (Ko¨ppen Group H) stretching from the southwest to the north part of the country, Group BS climate in the north near the coast to the Caribbean Sea, Group Aw in the north and Group Af in the south and following the Pacific coastline in the northwest. Extreme places identified are:
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
10 18 26 17 26 35 60 72 85 1 9 16
Area west of La Chorrera, department Amazonas, in the south; Tarapaca´, located in the tropical rain forest north of Letı´cia, department Amazonas, at the Amazonas river in the south at the border to Brazil and Peru; Sahagu´n, department Co´rdoba; Sa. de Chiribiquete, department Guaviare, in the south.
The hottest place identified in Colombia is in the northwest at 75.48W, 10.08N, south of Arjona, at the coast to the Caribbean Sea southeast of Cartagena, department Bolı´var. Mean temperatures increase to more than 308C in January, February and March, and are never lower than 258C (see Fig. 7). This is also one of the most humid places found in Colombia.
Table 8. Climatic Data for Guyana
City
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Georgetown (capital)
25.4
25.6
17
28.26
86.9
66.5
Skeldon
25.8
26.0
15
28.58
85.9
67.3
Raad
26.4
26.7
13
25.35
73.7
59.7
DOI 10.1002/jps
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
2 5 13 3 4 12 9 17 26
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
956
ZAHN ET AL.
Figure 11. Daily temperature fluctuations—Puerto Curaray, Peru.
The most humid place in Colombia is in the northwest at 76.68W, 8.98N near San Juan (department Antioquia), at the coast to the Caribbean Sea. The mean maximum partial pressure values increase to more than 31 hPa from April to August, and are always higher than 28 hPa with minimum values in January and February (see Fig. 8). The climatic parameters for Colombia are summarised in Table 6.
Figure 12. Daily partial vapour pressure fluctuations—Northeast corner, Peru.
Ecuador Group Aw climate can be found along the coast to the Pacific Ocean in the west, and Group Af in the east part of the country. The capital Quito offers a moderate climate due to the fact that the city is located at 2,800 m above sea level. Mean temperatures fluctuate between 158C and 258C, and values for partial vapour pressure never increase to more than 20 hPa.
Table 9. Climatic Data for Peru
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Lima (capital)
15.2
16.2
85
11.34
65.7
26.7
Nauta, Loreto
25.9
26.3
14
28.14
84.3
66.3
Northeast corner, Loreto
25.4
25.7
17
28.61
87.9
67.4
Puerto Curaray, Loreto
26.5
27.0
11
25.73
74.2
60.6
Puerto Salvatierra, Loreto
26.5
26.9
12
25.05
72.4
59.0
River Tamaya, Ucayali
24.8
25.1
20
27.54
88.2
64.8
San Antonio, Loreto
26.5
26.9
11
26.16
75.5
61.6
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
143 162 181 2 6 13 4 4 11 7 16 24 10 19 27 0 8 16 6 14 22
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
The highest mean temperatures have been found near Sarayacu, province Pastaza, east of Riobamba. Temperatures here fluctuate between 228C and 328C, and there is not a single day throughout the year when the mean maximum temperature drops below 308C (see Fig. 9). The most humid place in Ecuador is the Yasunı´ National Park, province Orellana, in the east close to the border to Peru (see Fig. 10). The place with the highest mean relative humidity is Jipijapa, province Manabı´, south of Manta at the east coast to the Pacific. The mean temperatures fluctuate by not more than 28C between day and night, but there is a difference between summer and winter: The mean minimum temperatures can be detected at night in August at just above 218C, and the mean maximum temperatures during the day in March at almost 278C (see Tab. 7). Guyana The climate in Guyana is in Group Am. There is a Group Aw climate in the southwest part of the country. The hottest place is the savanna in the southwest of the country, southeast of Raad at 59.68W, 2.28N, with mean maximum temperatures reaching 318C (during the day in March and again in September and October), and mean minimum temperatures above 238C (during the night in July). The most humid place can be found in the northeast part of Guyana near the North Atlantic coast, southwest of the small city Skeldon at 57.38W, 5.68N. The mean partial vapour pressure values never drop below 26 hPa (in February), but reach peaks of more than 30 hPa (in July). Details are listed in Table 8. Peru The western coastal plain facing the Pacific Ocean belongs to Group BW, the Andes in the centre to
Table 10.
Figure 13. Daily temperature fluctuations—Paramaribo, Suriname.
Group H and the Amazon Basin in the northeast part to Group Af. The hottest place in Peru is west of Puerto Curaray, department Loreto (74.38W, 2.48S). The daily mean maximum temperatures never drop below 298C, and the lowest mean temperature found is 22.68C (at night in July) (see Fig. 11). The most humid place has been identified as a place in the northeast corner of Peru, department Loreto, north of the Amazon river close to the border to Colombia (70.98W, 2.48S) (see Fig. 12). The highest mean relative humidity has been found at the river Tamaya in the department Ucayali, at the border to Brazil (73.18W, 9.08S). This is due to high partial vapour pressure values at lower temperatures. Other extreme places analysed are Puerto Salvatierra, department Loreto (76.48W, 3.68S), and San Antonio, Loreto (74.38W, 3.68S). Both are almost as hot as Puerto Curaray. It is very humid east of Nauta, Loreto, near the river Yavarı´ Mirim close to the border to Brazil (73.18W, 4.68S). Lima, the capital of Peru, is located at the Pacific coast in a moderate climatic zone with mean
Climatic Data for Suriname
City
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Paramaribo (capital)
26.1
26.3
14
28.22
83.3
66.4
Totness
26.1
26.3
14
28.08
83.0
66.1
DOI 10.1002/jps
957
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75
2 5 13 2 6 13
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
958
ZAHN ET AL.
Figure 15. Daily partial vapour pressure fluctuations—Capata´rida, Venezuela.
Figure 14. Daily temperature fluctuations—Puerto Piritu, Venezuela.
temperatures fluctuating between 118C and 238C, and mean PD values below 10 hPa from June to October. Details are listed in Table 9.
equal high levels throughout the country but more pronounced towards the Atlantic coast (see Tab. 10).
Suriname
Venezuela
The dominating climate is in Group Am. The hottest place is southeast of Paramaribo, the capital of Suriname, at 55.18W, 5.68N, with minimum mean temperatures of less than 248C during the nights in January to March, and maximum mean temperatures of more than 308C during the days in September and October (see Fig. 13). Partial vapour pressures are at
The northern half of Venzuela belongs to Group BS, the coast to the Caribbean Sea in the north to Group BW, the central plains and the Guiana Highlands in the southeast to Group Aw. One of the extreme places is at the southwest corner of the country at 66.48W, 1.18N, southeast of El Carmen (Amazonas State), near the borders to Brazil and Colombia, with the highest mean
Table 11. Climatic Data for Venezuela
City, Province
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
Capata´rida, Falco´n
27.3
27.4
10
29.24
80.7
68.8
Caracas (capital)
26.3
26.6
13
24.96
73.0
58.8
El Carmen, Amazonas
25.2
25.4
18
28.60
89.3
67.3
El Sama´n, Apure
27.0
27.4
9
26.82
75.0
63.1
Paraguaipoa, Zulia
27.3
27.6
9
27.61
76.2
65.0
Puerto Pirito, Anzoa´tegui
27.3
27.7
8
25.13
69.0
59.2
Santa Marı´a de Iripe, Gua´rico
27.2
27.6
9
24.75
68.6
58.3
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
RH [%]
RH [%] at 308C
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75 30/65 30/70 30/75
6 2 9 11 19 28 4 4 11 3 11 19 0 8 15 10 18 27 12 20 29
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
Table 12.
959
Climatic Data for China
City, Provinces
T [8C]
MKT [8C]
Hong Kong (HKG)
23.4
24.4
Macau (MAC)
23.0
24.1
Mengla, Yunnan
21.1
22.1
Nanchang, Jiangxi
18.0
21.5
Nanchong, Sichuan
18.1
20.9
Sanya, Hainan
26.0
26.4
Shanghai, Shanghai
15.8
19.0
YT [%] T ¼ 258C T ¼ 308C 3 23 4 24 13 36 16 40 19 43 5 14 32 58
PD [hPa]
RH [%]
22.21
77.4
22.12
78.9
19.34
77.3
15.73
76.0
15.29
73.7
27.16
80.7
13.96
77.8
RH [%] at 258C at 308C
Testing Condition [8C/% RH]
70.1 52.1 69.8 52.1 61.0 45.5 49.6 37.0 48.2 36.0 85.7 64.0 44.0 32.9
25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65
YPD [%] 14 25 14 25 2 43 21 76 24 81 30 2 36 98
relative humidity due to high partial vapour pressures at lower temperatures. A hot climate has been found near El Sama´n (Apure State), at 68.68W, 7.88N and also at 65.28W, 8.98N, northeast of Santa Marı´a de Iripe (Gua´rico State). The climate south of Paraguaipoa (Zulia State), at 71.9 W, 11.2 N, at the coast to the Golfo de Venezuela, in the northwest corner of the country, is also of interest. Caracas, the capital of Venezuela, northeast of 67.58W, 10.08N, at the coast to the Caribbean Sea, presents a climate which is less humid than the extreme places identified. The hottest place is in the area of 65.28W, 10.08N, which is west of Puerto Piritu (Anzoa´tegui State) at the coast to the Caribbean Sea. There is not one single day when the mean maximum temperature is below 308C, while mean minimum temperatures never decrease to less than 248C (during the night in January and February) (see Fig. 14).
The place with the highest mean partial vapour pressures in Venezuela is at 70.88W, 11.28N, west of Capata´rida (Falco´n State), in the northwest part of the country at the coast to the Golfo de Venezuela. Mean maximum water vapour pressures reach almost 32 hPa in September and October, and mean minimum values can be found above 26 hPa in January and February (see Fig. 15). The climatic parameters for Venezuela are listed in Table 11.
Figure 16. Daily temperature fluctuations—Sanya, Hainan, China.
Figure 17. Daily partial vapour pressure fluctuations—Sanya, Hainan, China.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
CHINA Mainland China is extremely diverse from a climatic point of view: from tropical parts in the south (Group Am) to regions with severe dry winters (Group Dw) in the northeast, and humid subtropical regions (Group Cfa) in between.
960
ZAHN ET AL.
Figure 19. Daily partial vapour pressure fluctuations—Cape Saint Lucia, Natal, South Africa.
Figure 18. Daily temperature fluctuations—Cape Saint Lucia, Natal, South Africa.
Shanghai at the east coast, Macau and Hong Kong in the south east and the city located in the most southern corner of the mainland, Mengla close to the border to Laos, have been selected. Another city in the south is Sanya on the southern coast of the island Hainan. In addition to these cities, two cities identified as hot and humid ‘spots’ in the centre of the land mass have been analysed, Nanchang in Jiangxi, and Nanchong in Sichuan. The results of the calculations are presented in Table 12 for both testing at 258C/60% RH and 308C/ 65% RH. As can be concluded from the table, the most loading place in China is Sanya on the island Hainan in the most southern part of the country.
Mean maximum temperatures reach more than 308C in four consecutive months (May to August), and mean minimum temperatures never drop below 208C (in January, see Fig. 16). The southern part of this island belongs to Ko¨ppen Group Am, the northern part to Group Aw. The maximum PD values increase to more than 33.0 hPa during 3 months (June to August, see Fig. 17). To calculate the adequate relative humidity for long-term stability testing, the mean partial water vapour pressure calculated for Sanya (27.16 hPa) is used at the standard testing temperature 308C to get 64.0% RH. Testing at 308C would include a safety margin of 14% added to the MKT, and
Table 13. Climatic Data for South Africa
T [8C]
MKT [8C]
Cape Saint Lucia, KwaZulu-Natal Cape Town, Western Cape
22.3
22.6
16.7
17.0
De Aar, Northern Cape
16.7
19.2
Durban, KwaZulu-Natal
20.6
21.0
Kimberley, Northern Cape
17.6
20.3
Little Namaland, Northern Cape
21.4
23.6
Musina (Messina), Limpopo
21.5
22.8
Port Elizabeth, Eastern Cape
18.8
19.1
Pretoria, Gauteng
17.5
19.2
City, Provinces
YT [%] T ¼ 258C T ¼ 308C
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
11 33 47 77 30 57 19 43 23 48 6 27 10 32 31 57 30 56
PD [hPa]
RH [%]
20.66
76.9
14.96
78.8
8.26
43.5
19.28
79.4
9.76
48.4
9.65
37.8
15.73
61.3
16.27
75.1
10.70
53.4
RH [%] at 258C at 308C
Testing Condition [8C/% RH]
YPD [%]
65.2 48.6 47.2 35.2 26.1 19.4 60.8 45.4 30.8 23.0 30.4 22.7 49.6 37.0 51.3 38.3 33.8 25.2
25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65
8 34 27 85 130 234 1 43 95 183 97 186 21 76 17 70 78 158
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
961
Figure 21. Daily partial vapour pressure fluctuations—Mbandaka, Dem. Rep. Congo.
Figure 20. Daily temperature fluctuations—Dar es Salaam, Tanzania.
SOUTH AFRICA testing at 65% RH would include a safety margin of 2% for PD. Sanya, however, presents an extreme climate compared to the other parts of the country.
Table 14.
Climatic Data for SADC (Without South Africa)
Country: City
T [8C]
MKT [8C]
Angola: Luanda
25.1
25.7
Botswana: Gaborone
19.5
21.5
Dem Rep Congo: Boma
24.6
24.8
Kinshasa
24.1
24.4
Mbandaka
24.5
24.8
Lesotho: Maseru
13.8
15.9
Malawi: Lilongwe
21.9
22.7
Mauritius
24.5
24.7
Mozambique: Maputo
22.8
23.6
Namibia: Windhoek
21.5
23.3
Swaziland: Mbabane
18.9
20.1
Tanzania: Dar es Salaam
26.2
26.6
Zambia: Lusaka
21.7
22.7
Zimbabwe: Harare
19.2
20.2
DOI 10.1002/jps
There is only a small part of South Africa presenting a Group Cfa climate, that is the Durban area near the Indian Ocean. The most humid place identified in South Africa is Cape
YT [%] T ¼ 258C T ¼ 308C 3 17 16 40 1 21 3 23 1 21 58 89 10 32 1 22 6 27 7 29 24 49 6 13 10 32 24 48
PD [hPa]
RH [%]
23.48
73.6
11.84
52.2
26.25
84.9
24.14
80.6
26.91
87.5
8.01
50.9
16.92
64.4
23.39
76.2
19.22
69.1
7.27
28.4
16.01
73.1
25.26
74.2
15.27
58.9
13.63
61.4
RH [%] at 258C at 308C
Testing Condition [8C/% RH]
74.1 55.3 37.4 27.9
25/60 30/65 25/60 30/65
19 18 61 133
82.8 61.8 76.2 56.8 84.9 63.4 25.3 18.9 53.4 39.8 73.8 55.1 60.6 45.3 22.9 17.1 50.5 37.7 79.7 59.5 48.2 36.0 43.0 32.1
25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65 25/60 30/65
28 5 21 14 29 3 137 245 12 63 19 18 1 44 162 280 19 72 25 9 25 81 40 103
YPD [%]
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
962
ZAHN ET AL.
Table 15. Climatic Data for Cuba
City
T [8C]
MKT [8C]
YT [%] T ¼ 308 C
PD [hPa]
RH [%]
RH [%] at 308C
Testing Condition [8C/% RH]
YPD [%]
Camagu¨ey Santiago de Cuba West of Santa Clara
26.1 26.6 25.3
26.4 26.9 25.7
14 12 17
24.69 25.98 23.40
72.9 74.6 72.6
58.1 61.2 55.1
30/65 30/65 30/65
12 6 18
Saint Lucia at the west coast north of Durban. Mean maximum temperatures can go up to 278C at noon in February, combined with a mean maximum partial vapour pressure of almost 26 hPa. Mean minimum temperatures decrease just below 188C at night in July (see Figs. 18 and 19). In the centre of South Africa, the Group BS is dominating. Pretoria and Johannesburg are located there, while Cape Town is located in a Group Bsk climate. At the south coast, South Africa presents a Group Cfb climate (see Tab. 13).
SADC Tropical Climatic Group Af is to be found north and south of the equator, surrounded by Group Aw regions. The hottest place in SADC identified so far is Dar es Salaam (Group Aw) in Tanzania (see Fig. 20). Mean maximum temperatures increase to 31.68C in February, and never decrease below 218C. The most appropriate temperature for longterm stability testing of medicinal products to be marketed in Tanzania and SADC is 308C. That value includes a safety margin of 13% added to the highest MKT calculated for Dar es Salaam. The most humid place is Mbandaka (Group Af): Values for partial water vapour pressure increase
to almost 29.0 hPa in May, and decrease to a minimum of 24.9 hPa in July (see Fig. 21). The mean PD value of 26.91 hPa calculated for Mbandaka at the testing temperature of 308C would result in a relative humidity of 63.4% RH. Testing at 308C/65% RH would include a safety margin of 3% for PD (see Tab. 14).
CARIBBEAN ISLANDS All of the northwestern Caribbean Islands are in Ko¨ppen Group Aw, the southeastern islands belong to Group Af. Cuba Three representative places have been selected, one southeast of Havanna, west of Santa Clara, another in the centre of the eastern part near Camagu¨ey and the third near Santiago de Cuba at the southeast coast of the island. The island belongs to Group Aw. For long-term stability testing of medicinal products to be marketed in Cuba, 308C is appropriate. This value includes a safety margin of 12% added to the highest MKT calculated for Cuba (see Tab. 15). The mean PD value of 25.98 hPa calculated for Santiago de Cuba at the testing temperature of
Figure 22. Daily temperature fluctuations—Curac¸ao, The Netherlands Antilles.
Figure 23. Daily partial vapour pressure fluctuations—Curac¸ao, The Netherlands Antilles.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
Table 16.
963
Climatic Data for Curac¸ao and Puerto Rico
Island
T [8C]
MKT [8C]
YT [%] T ¼ 308C
PD [hPa]
RH [%]
RH [%] at 308C
Curac¸ao
27.5
27.6
9
29.46
80.2
69.4
Puerto Rico
26.8
26.9
12
27.51
78.2
64.8
308C would result in a relative humidity of 61.2% RH. Testing at 308C/65% RH would include a safety margin of 6% for PD (see Tab. 15). Curac¸ao The island’s climate is similar but more extreme than the other Caribbean Islands. Mean temperatures can go up to 29.88C in October, and never decrease to less than 25.48C in February (see Fig. 22). Mean partial pressures reach a maximum of 32.27 hPa in September (see Fig. 23 and Tab. 16). Puerto Rico The climate is characterised by high temperatures throughout the year, that is, fluctuation between 24.38C in February and 29.48C in September, and high humidity, that is, up to 31.14 hPa. It belongs to Ko¨ppen Group Aw. Testing at 308C/65% RH would include a safety margin of 12% for the MKT, but 0% for PD (see Tab. 16).
Testing Condition [8C/% RH]
YPD [%]
30/65 30/70 30/75 30/65 30/70 30/75
6 1 8 0 8 16
DISCUSSION As a first step to develop an adequate long-term stability testing condition for hot and humid countries and regions, meteorological data based on monthly mean values of temperatures and dewpoints led to the evaluation of a climate of a particular country or region. The next step was the identification of the hottest and/or most humid place in a country assuming that a medicinal product can be shipped and stored everywhere within the boundaries of a country where the product is allowed to be marketed. This concept can be extended in principle to a group of associated countries, for example ASEAN. As the hottest and most humid place is representing the upper end of potential climatic conditions instead of an average of all climates, there will be a difference between the testing and the climatic conditions for places with lower humidities. This difference can be calculated by applying the safety margin Y for temperature and partial water vapour pressure PD. Assessing the results of these calculations facilitates the assignment of a country to either Climatic Zone IVA or IVB.
Table 17. Long-Term Testing Conditions for ASEAN Member Countries
Country Brunei Cambodia Indonesia Laos Malaysia Myanmar Philippines Singapore Thailand Vietnam
DOI 10.1002/jps
308C/65% RH, 308C/70% 308C/75% RH, CZ IVA RH CZ IVB þ þ þ þ
Country þ
þ
þ
Table 18. Long-Term Testing Conditions for South American Countries
þ þ þ
Bolivia Brazil Colombia Ecuador Guyana Peru Suriname Venezuela
308C/65% RH, 308C/75% RH, CZ IVA 308C/70% RH CZ IVB þ þ
þ þ
þ þ þ þ
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
964
ZAHN ET AL.
Table 19. Long-Term Testing Conditions for China Country
Country
þ þ þ
CONCLUSION As a result of this risk assessment, an appropriate long-term stability testing condition is proposed for each selected country (see Tabs. 17–21). In addition to the long-term testing condition 258C/ 60% RH for markets in Climatic Zone II, the hot and humid countries are categorized as either Table 20. Long-Term Testing Conditions for SADC Member Countries 258C/60% RH, CZ II þ þ þ þ þ þ þ
308C/65% RH, CZ IVA
Cuba Curac¸ao Puerto Rico
An adequate long-term testing condition should be established by balancing the need to provide the patients with stable medicinal products with the need to reduce costs of goods for industry. If the testing conditions were to be too stressful, the consequence would be unnecessary protective packaging (e.g. double aluminium blister), a reduced shelf-life (e.g. 2 years instead of 3 years) or too restrictive storage recommendations (store below 258C). Careful analysis of the climate in all tropical and subtropical parts of the world, and calculation of the relative humidity at the standard testing temperature 308C offer the possibility to evaluate the ‘‘probability of failure’’ of a particular product in a particular market, that is the risk for insufficient stability.
Angola Botswana Dem. Rep. Congo Lesotho Malawi Mauritius Mozambique Namibia South Africa Swaziland Tanzania Zambia Zimbabwe
Long-Term Testing Conditions for Caribbean
308C/65% RH, CZ IVA
China (Mainland) Hong Kong Macau
Country
Table 21. Islands
308C/65% RH, CZ IVA þ þ þ þ þ þ
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006
þ
308C/70% RH þ þ
belonging to category CZ IVA, for which testing at 308C/65% RH would be sufficient, or to CZ IVB, if testing at 308C/75% RH were required. The long-term testing condition 308C/70% RH has been kept in these tables to facilitate the risk assessment for those products, for which 308C/75% RH data are not yet available. This would provide manufacturers with more flexibility in the transition period when testing at 308C/70% RH could eventually be replaced by the condition 308C/75% RH.
REFERENCES 1. WHO. 1996. Guidelines for stability testing of pharmaceutical products containing well established drug substances in conventional dosage forms. WHO Technical Report Series 863, Annex 5. 2. WHO. 2003. Report of the 37th meeting of the WHO Expert Committee on Specifications for Pharmaceutical Preparations, Geneva, 22–26 October 2001. WHO Technical Report Series 908, page 13. 3. ICH Q1F. 2003. Stability data package for registration applications in Climatic Zones III and IV. http://www.ich.org/LOB/media/MEDIA416.pdf 4. Slamet LS, et al. 2004. Stability testing for hot and humid climates. WHO Drug Inf 18:113– 116. 5. WHO. A WHO consultation of stability studies in a global environment was held in Geneva on 13–14 December 2004. 6. Resolution RE No. 1, dated 29 July 2005, published in the Official Gazette of the Union, Supplement to No. 146- Section 1, Brası´lia- DF, on 1 August 2005. 7. WHO. 40th WHO Expert Committee on Specifications for Pharmaceutical Preparations, Geneva, 24–28 October 2005. ¨ ber eine fu¨r die Haltbarkeit 8. Schumacher P. 1972. U von Arzneimitteln maßgebliche Klimaeinteilung. Pharm Ind 34:481–483. 9. Grimm W. 1986. Storage conditions for stability testing (Part 2). Drugs made in Germany 29:39–47. 10. Grimm W. 1998. Extension of the International Conference on Harmonization Tripartite Guideline DOI 10.1002/jps
STABILITY TESTING CONDITIONS FOR TROPICAL COUNTRIES
965
for stability testing of new drug substances and products to countries of Climatic Zones III and IV. Drug Dev Ind Pharm 24:313–325. 11. Ref 1, Table 2 and Table 3, page 70. 12. The United States Pharmacopeia (USP) 28 (2005), General Chapter 1150 Pharmaceutical Stability. 13. Uppala SM, et al. 2005. The ERA-40 Reanalysis. Q J R Meteorological Soc 131:2961–3012.
14. Hardy R. 1998. ITS-90 Formulations for vapor pressure, frostpoint temperature, dewpoint temperature, and enhancement factors. In Proceedings of the Third International Symposium on Humidity & Moisture, Teddington, London. 15. Haynes JD. 1971. Worldwide virtual temperatures for product stability testing. J Pharm Sci 60:927– 929.
DOI 10.1002/jps
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 5, MAY 2006