- Email: [email protected]
recombinant-B-subunit cholera vaccine in Peruvian military recruits
Protective efficacy of oral whole-cell/recombinant-B-subunit cholera vaccine in Peruvian military recruits
Summary The cholera epidemic in South America has reinforced the need for safe and effective oral vaccines. In a randomised, double-blind, placebo-controlled efficacy trial among 1563 Peruvian military recruits we have investigated the protective efficacy of an oral inactivated wholecell/recombinant-B-subunit (WC/rBS) cholera vaccine. Participants were given two oral doses of cholera vaccine or Escherichia coli K12 placebo, with an interval of 7-14
days. 1426 (91%) subjects received the two prescribed doses and were followed up for a mean of 18 weeks (median 21 weeks). After vaccination, Vibrio cholerae O1 El Tor Ogawa was isolated from 17 subjects with diarrhoea. 16 of the cholera cases occurred 2 weeks or longer after the second dose of vaccine (14 placebo recipients, 2 vaccinees). We also detected 14 symptomless infections (11 [7 placebo recipients, 4 vaccinees]) 2 weeks or longer after the second dose. The vaccine had significant protective efficacy against cholera (86% [95% Cl 37-97], p<0·01) but not against symptomless infection (42% [-96 to 85]). All cholera cases were in people of blood group O, who made up 76% of the study population (p<0·01). Two doses of WC/rBS vaccine, given 1 to 2 weeks apart, provide rapid, short-term protection against symptomatic cholera in adult South Americans, who are predominantly of blood group O. Long-term efficacy studies in Peruvian adults and children are under way.
Division of Preventive Medicine, Walter Reed Army Institute of Research, Washington DC, USA (J L Sanchez MD); United States Navy Medical Research Institute Detachment, Lima, Peru (B Vasquez MD, R E Begue MD, R Meza BS, G Castellares MD, C Cabezas MD, D M Watts PhD); Department of Medical Microbiology and Immunology, University of Göteborg, Sweden (A-M Svennerholm MD); and Division of Communicable Diseases and Immunology, Walter Reed Army Institute of Research, Washington, DC, USA (J C Sadoff MD, D N Taylor MD)
Correspondence to: Dr Jose L Sanchez, US Army Medical Research Unit-Brazil, American Consulate-Rio, Unit 3501, APO AA 34030, Rio de Janeiro, Brazil
Introduction The cholera epidemic in Latin (Central and South) America has had enormous medical, economic, and social consequences. In 3 years nearly a million cases and more than 8000 deaths from cholera have been reported.’ In Peru, an estimated 495 million dollars were lost in 1991 in the areas of tourism, fishing, exports, and health.2 Although many lives have been saved with oral rehydration, many patients with severe cholera require initial intravenous rehydration. The cost of intravenous fluids is prohibitively expensive in the developing world. Improvement of the potable water supply and sanitation in Latin America is recognised as an important priority for the long-term reduction of enteric diseases.3 However, the necessary improvements in the infrastructure will require billions of dollars and decades to complete.’I The recent cholera epidemic has renewed interest in the development of safe and effective orally administered cholera vaccines. An oral cholera vaccine consisting of killed cholera organisms plus the B subunit of cholera toxin was well tolerated and provided significant protection against cholera and enterotoxigenic Escherichia coli (ETEC) in Bangladesh.4-7 This vaccine provided 85% efficacy against cholera in the first 6 months and 50% cumulative efficacy over 3 years. Although the Bangladesh trial was designed to test a regimen of three doses given with 6-week intervals, subgroup analyses suggested that a two-dose regimen was equally effective.6 A less expensive formulation of this vaccine has been developed with a recombinant form of the B subunit (rBS).8 This formulation (WC/rBS) was found to be safe and immunogenic in North American and Swedish adults and Peruvians of all ages.9-11 In the Bangladesh trial the vaccine had significantly lower than average efficacy among children aged 2-5 years (presumably because of a lack of previous immunity), among people infected with the El Tor biotype6 and among people of blood group 0.12 These factors could lower the efficacy of this vaccine in Latin America, since the outbreak was caused entirely by the El Tor biotype, the population had no previous immunity to cholera, and about 75% of people are of blood group 0. Because of these factors the WC/rBS vaccines had to be studied in controlled clinical trials before it could be recommended for widespread use.
Patients and methods Each dose of WC/rBS vaccine (SBL Vaccin AB, Stockholm, Sweden) consists of 1 mg recombinantly produced cholera toxin B subunit and 2-5X10"’ of each of the following Vibrio cholerae
1273
in 3 mL phosphate-buffered saline: heatkilled classical Inaba (strain Cairo 48), heat-killed classical Ogawa (strain Cairo 50), formalin-treated El Tor Inaba (strain Phil 6973), and formalin-treated classical Ogawa (strain Cairo 50). Each 3 mL dose of vaccine was administered with a sodium bicarbonate buffer (Samarin, Cederroths Nordic AB, Upplands Vasby, Sweden) in 150 mL water containing 3-8 g sodium bicarbonate, 2-4 g tartaric acid, and 0-81 g citric acid. The placebo consisted of a suspension of heat-inactivated E coli K12 strain (SBL Vaccin AB) in a concentration that matched the turbidity and appearance of the vaccine preparation. The agents were dispensed in 300 mL bottles and stored at 4-8°C. Each bottle was identified with a unique number; vaccine and placebo
organisms suspended
preparations were pre-coded. Volunteers were enrolled in January and March, 1994, from three military training centres near Lima, Peru (CITEN, IMAP, and FAP). To be eligible, volunteers had to be 17-65 years old, to be at the centre for a minimum of 3 months, and to give written informed consent. No subject had received cholera vaccine previously. The study was approved by the Institutional Review Board of the Peruvian Navy Surgeon General, the Walter Reed Army Institute of Research Human Use Review Committee, and the Human Subjects Research Review Board, Office of the Army Surgeon General. The main objective of this study was to assess the efficacy of the vaccine against cholera. The only question of interest was whether the vaccine was more (and not less) protective than placebo. Therefore, the study was designed to detect a vaccine efficacy of 70% or more at p<0-025 (one-tailed) with a power of 0-8. On the basis of previous experience of cholera outbreaks at the participating military bases we assumed that the cumulative frequency of cholera would be 2-5% during the 3 months of basic training, therefore 457to 1165subjects per group would be required. Randomisation was done in blocks of 10 (5 to vaccine, 5 to placebo) to ensure equal study groups. Adverse events were not actively monitored in this study and no samples were taken for serology. Passive surveillance for diarrhoea was carried out at the medical clinics serving the training centres. Doctors from the Peruvian armed forces attended all diarrhoea cases on a 24 h basis. Subjects in the study were frequently reminded to report any episode of diarrhoea. For any patient with diarrhoea who was taking part in the study, the information contained on his study identification card was recorded at presentation. Within 48 h his identity was confirmed against a master list of study participants kept at the US Navy Medical Research Institute Detachment (NAMRID). Information on non-participants with diarrhoea was collected but not used for efficacy analyses. Clinical information was collected on standard data forms and a stool-impregnated rectal swab was collected and stored in Cary-Blair transport medium. Samples were transported to NAMRID and processed within 48 h of collection. Conventional techniques were used to isolate V cholerae 01 and to find out the biotype and serotype of each isolate.When outbreaks occurred, epidemiological information and rectal swabs were also obtained from symptom-free people in the recruit groups most affected. Such cross-sectional surveys were carried out four times (Feb 3 at CITEN, and Feb 8, March 29, and April 6 at IMAP). The incidence of cholera in the study group had exceeded 2% by June 1, 1994. Since no cases of cholera had occurred in the bases for 6 weeks, we decided to end the study on June 15, 1994. Data files were verified and edited, including determination of case and vaccine status, before the study code was broken.The cumulative time-in-study (in person-weeks) of those receiving two correct doses of each agent was recorded and used as the denominator to calculate incidence rates for each study group. Subjects lost to follow-up after the second dose were assumed to contribute half of the period to the denominator in the analysis. According to our protocol, vaccination was defined as the ingestion of two full doses of vaccine or placebo given 7-14 days apart. A case of cholera was defined as non-bloody diarrhoea (three or more loose stools in a 24 h period) with onset 2 or more weeks after the second dose, associated with the isolation of
1274
RR=rate ratio, weeks.
PE=protective efficacy.
Incidence rate ?
cases
per 1000 person-
*p<0’01 (one-tailed) for comparison of vaccine versus placebo Table : Protective efficacy of vaccine V cholerae 01 1 from a faecal sample. Severe cholera was defined as diarrhoea of any duration plus two other symptoms or objective signs of dehydration (lethargy, severe thirst, rapid and feeble pulse, tachypnoea, decreased skin turgor, sunken eyes, low urine output).ABO blood group tests were done by the Lee-Vincent 15 method on fingerstick samples. Vaccine protective efficacy was estimated as: rotectlve e Protective efficacy= lcacy=
Attack rate in unvaccinated-attack Attack rate m unvaccinated
rate m
vaccmated
X 100
95% CI on protective efficacy were calculated by accepted methods for field trials of vaccines. Incidence rates, rate ratios (incidence rate in vaccinated/incidence rate in unvaccinated), and 95% CI on rate ratios were also calculated. Proportions were compared by the X2 or Fisher’s test (two-tailed).
Results The
study population consisted of 1563 Peruvian men (military recruits and students) with an average age of 19 years (range 16-45). The subjects were randomly allocated vaccine (781) or placebo (782). 1426 (91%) received two doses of vaccine or placebo. Surveillance continued for 12 596 person-weeks among vaccinees and 12 794 person-weeks for placebo recipients. Losses to follow-up were minimal and similar in the two groups (22 [3%] of 710 vaccine recipients and 16 [2%] of 716 placebo recipients). The two-dose vaccination regimen was given within 2 weeks of the recruits’ arrival on-base. Of the 1426 subjects, 97% (1379) received two doses with an interval of 7-11 days between doses and 3% (47) received the two doses 27 days apart. Between Jan 4 and June 15, 1994, 383 (25%) of the study subjects went to the clinic with diarrhoea and samples were taken. Cases started to occur 2-3 weeks after the second dose was given at the CITEN site (late January, 1994). Two outbreaks occurred at the IMAP site within 2-4 weeks of vaccination (early February and early April, 1994). These two study sites accounted for all but 6 of the diarrhoea cases reported among study subjects. V cholerae 01was isolated from 38 (10%) of diarrhoea cases among study participants. The overall attack rate for cholera in the study population was therefore 2-4% (38 cases among 1563 participants). A further 20 cholera cases were identified among non-participants (estimated population 6000). V cholerae El Tor Ogawa was isolated from 54 subjects and V cholerae El Tor Inaba from 4. Of the 38 cases among study participants, 16 (42%) satisfied the criteria for inclusion in the efficacy analysis (two-dose recipients, 2 or more weeks after the second dose). Of the 22 cases not included in the efficacy analysis, 21 occurred before the second dose was given and only 1 case (in a placebo recipient) occurred shortly after the second dose. In addition, 14 symptomless infections were diagnosed among study subjects (all caused by V cholerae El Tor Ogawa). 11 (79%) of these occurred among two-dose
recipients.
after the second dose when the interval between doses is 7-14 days. Thus, protection can be provided within 3 weeks of the first dose. This short interval could make the vaccine a useful public health intervention for the prevention of cholera epidemics. We also found that WC/rBS cholera vaccine was very effective in protecting Peruvian adults against El Tor cholera. The protection achieved in our study was similar to that shown by Clemens et al4 in the first 6 months of a large-scale efficacy study in Bangladesh (77% [95% CI 25-92]). Protection was achieved after only two doses given in a short time. A regimen of this nature is much simpler to implement than one with two or three doses given 6 weeks apart. These data support observations in Swedish adults that equivalent immunogenicity can be attained when two doses of this vaccine are given with intervals of 1-6 weeks. 17 It also supports previous findings from Bangladesh that two doses were as effective as three in preventing cholera after 3 years of surveillance.6 This study is the first demonstration that the vaccine is effective in Latin America, where El Tor is the only biotype and where the population is predominantly of blood group 0. In Bangladesh’2 there was a significantly lower protective efficacy among subjects of blood group 0 than among other subjects (40 vs 76%). Others have noted the predisposition for severe cholera among hospital Ínpatients2,18.19 as well as in volunteers2O of blood group 0. Our study supports those observations. The cholera attack rates (2-3%) seen during the military outbreaks are about ten times higher than those being experienced among civilian populations in Peru.2>2’ The epidemic setting observed in this trial is similar to that observed in other military forces and refugee populations and to conditions at the onset of the cholera epidemic in Peru in 1991. It is well known that the epidemiological features of epidemic cholera differ from those where cholera has become endemic.22 The pattern of outbreaks with high attack rates among adults seen in epidemic cholera, when contaminated food items and water are often implicated in the transmission of V cholerae, more closely resembles the conditions in our
shortly
Figure: Cumuiative frequency of cholera among study groups p
2 of the 16 cases of cholera in two-dose recipients occurred among vaccinees and 14 among placebo recipients (table). The WC/rBS vaccine therefore conferred significant protection against symptomatic cholera (protective efficacy 86% [95% CI 37-97%]). Protection was evident 2 weeks after the second dose and was maintained for the rest of the study period (figure) There were 4 symptomless cholera infections in the vaccine group compared with 7 among placebo recipients. The protective efficacy against symptomless infection was not significant (42% [-96 to 83%]). Rates of diarrhoeal illness (non-cholera diarrhoea) 2 weeks or longer after the second dose were similar in vaccine and placebo groups. Similarly, there was no difference in diarrhoea incidence during the 7-13 days after the second dose. Further investigation of cholera infections among 780 placebo recipients was done to assess potential associations between blood group 0 and infection status, as well as severity of infection. 32 V cholerae 01 infections occurred among placebo recipients (infection rate 4-1%). In 22 of these cases, the patient had symptoms (cholera attack rate 2-8%). Symptomatic cholera occurred only in individuals of blood group 0 (22 of 558 with blood group 0 vs 0 of 180 of other blood types, p<0-01). Only 2 of the 22 cases were severe, and both were in people of blood group 0. For symptomless infections the rate of V cholerae 01infection was similar in 0 and non-0 blood groups (8 of 558 [1’4%] vs 2 of 180 [1-1%], respectively). Because very few cholera cases were detected in our study, no subgroup efficacy analyses could be done for blood group or severity of illness.
study. The short time between vaccination and the onset of cholera may have provided optimum conditions for the assessment of the protective efficacy of WC/rBS vaccine. Before the vaccine can be recommended for use on a wider scale, large-scale efficacy trials among choleraendemic populations in Lima and Arequipa must be completed. Studies of vaccine effectiveness and efficiency will also be required before this vaccine can be recommended for use in a public health programme.23
Discussion Since cholera appeared in Peru in 1991 there have been annual outbreaks of diarrhoea (and cholera) at military recruit training bases during the warm summer months (unpublished). These outbreaks have occurred soon after recruits arrive on-base, and confirmed cholera-attack rates are as high as 16% among affected units. This year we were able to vaccinate incoming recruits before outbreaks occurred. We found that if the second dose of vaccine could be given 2 weeks before a cholera outbreak
began, Only 1
References 1 2
3 4 5
1988; 158: 60-68.
it could case
weeks of the
provide significant protection (figure). occurred, in a placebo recipient, within 2 second dose. Immunity seems to develop
Pan American Health Organization. Cholera situation in the Americas. Epidemiol Bull 1994; 15: 13-16. Gotuzzo E, Cieza J, Estremadoyro L, Seas C. Cholera: lessons from the epidemic in Peru. Infect Dis Clin N Am 1994; 8: 183-205. Pan American Health Organization. Cholera vaccine evaluation. Epidemiol Bull 1991; 12: 5-10. Clemens JD, Sack DA, Harris JR, et al. Field trial of oral cholera vaccines in Bangladesh. Lancet 1986; ii: 124-27. Clemens JD, Harris JR, Sack DA, et al. Field trial of oral cholera vaccines in Bangladesh: results of one year of follow-up. J Infect Dis
6
Clemens JD, Sack DA, Harris JR, et al. Field trial of oral cholera vaccines in Bangladesh: results from three-year follow-up. Lancet 1990; 339: 270-73.
1275
7
Clemens JD, Stanton BF, Chakraborty J, et al. B subunit-whole cell and whole cell-only oral vaccines against cholera: studies on reactogenicity and immunogenicity. J Infect Dis 1987; 155: 79-85.
8
Holmgren J, Clemens J, Sack DA, Svennerholm A-M.
New cholera
vaccines. Vaccine 1989; 7: 94-96. Sanchez JL, Trofa A, Taylor DN, et al. Safety and immunogenicity of the oral, inactivated, whole cell plus recombinant B subunit of cholera toxin (WC/rBS) cholera vaccine in North American volunteers. J Infect Dis 1993; 167: 1446-49. 10 Jertborn M, Svennerholm A-M, Holmgren J. Safety and immunogenicity of an oral recombinant cholera B subunit-whole cell vaccine in Swedish volunteers. Vaccine 1992; 10: 130-32. 11 Begue RE, Castellares G, Ruiz R, et al. Community-based assessment of safety and immunogenicity of the whole cell plus recombinant B subunit (WC/rBS) oral cholera vaccine in Peru. Vaccine
9
(in press).
14
Kelly MT, Hickman-Brenner FW, Farmer III JJ. Vibrio. In: Balows A, Hausler WJ, Herrmann KL, Isenberg HD, Shadomy HJ, eds. Manual of clinical microbiology, 5th edition. Washington DC: American Society for Microbiology, 1991: 384-95. Bennish ML. Cholera: pathophysiology, clinical features, and treatment.
In: Wachsmuth IK, Blake PA, Olsvik O, eds. Vibrio cholerae
Society for Microbiology,
Garraty G. Blood group antigens and antibodies and pretransfusion compatibility testing. In: Rose NR, Conway de Macario E, Fahey JL, Friedman H, Penn GM, eds. Manual of clinical laboratory immunology, 4th edition. Washington DC: American Society for Microbiology, 1992: 308-19. 16 Orenstein WA, Bernier RH, Dondero TJ, et al. Field evaluation of vaccine efficacy. Bull WHO 1985; 63: 1055-68. 17 Jertborn M, Svennerholm A-M, Holmgren J. Evaluation of different
15
immunization schedules for oral cholera B subunit-whole cell vaccine in Swedish volunteers. Vaccine 1993; 11: 1007-12.
D, Paquio AS. ABO blood groups and cholera. Ann Hum Biol 1977; 4: 489-92. Chaudhuri A. Cholera and blood groups. Lancet 1977; ii: 404.
18 Barua 19
12 Clemens JD, Sack DA, Harris JR, et al. ABO blood groups and cholera: new observations on specificity of risk and modification of vaccine efficacy. J Infect Dis 1989; 159: 770-73.
13
and cholera. Washington DC: American 1994: 229-55.
20 Levine
MM, Nalin DR, Rennels MB, et al. Genetic cholera. Ann Hum Biol 1979; 6: 359-74.
21
susceptibility
to
Castellares G, Hayashi KE, et al. Diarrheal disease in Peru after the introduction of cholera. Am J Trop Med Hyg (in press).
Begue RE,
22 Glass 23
RI, Black RE. The epidemiology of cholera. In: Barua D, Greenough III WB, eds. Cholera. New York: Plenum, 1992: 129-54. Willems J, Sanders C. Cost-effectiveness and cost-benefit analyses of vaccines. J Infect Dis 1981; 144: 4486-93.
Teamwork
For
long time, general practice was epitomised by a partnership between patient and personal doctor, who worked single-handed. In the past few decades, however, medical care has become ever more complex and one consequence has been the emergence of primary-care teams.! Two characteristics make primary care unique in comparison with other sorts of medical practice-general practitioners provide "first" medical care for all categories of patients; and they provide care that is continuous over time. Potential members of the team should share, at least in essence, these approaches to patients. This paper reviews the development of the primary-care team from the perspective of the general practitioner-the partners in the team, the motives behind the formation of partnerships, and the consequences for individual care of patients. The
a
reference, which is
an important feature of a discipline; administrative staff usually work under supervision. In primary medical care, however, administrative and logistic support have become so important for day-to-day patient care that "protoprofessionalisation" is under way-a switch from administrative functions to direct involvement in patient care. 21 Various terms are used for this new function (practice nurse in the UK, practice assistant in the Netherlands), mirroring a mixed background of origin. Medical professionals in the community include occupational health officers and public health officers-for
whereas
primary-care team
Influence of health-service structure In health-care systems where there is a functional differentiation of tasks between primary and secondary care, most disciplines working in the community will have a primary-care orientation. Table 1 lists the professionals who offer unselected and ongoing care for broad categories of patients. The term discipline warrants some comment. The paramedical workers, psychologists, and social workers have an independent professional frame of
Department of General Practice and Social Medicine, University of Nijmegen, PO Box 9101,6500 HB Nijmegen, Netherlands (Prof C van Weel FRCGP)
1276
Table 1:
Primary care: the team
Summary The cholera epidemic in South America has reinforced the need for safe and effective oral vaccines. In a randomised, double-blind, placebo-controlled efficacy trial among 1563 Peruvian military recruits we have investigated the protective efficacy of an oral inactivated wholecell/recombinant-B-subunit (WC/rBS) cholera vaccine. Participants were given two oral doses of cholera vaccine or Escherichia coli K12 placebo, with an interval of 7-14
days. 1426 (91%) subjects received the two prescribed doses and were followed up for a mean of 18 weeks (median 21 weeks). After vaccination, Vibrio cholerae O1 El Tor Ogawa was isolated from 17 subjects with diarrhoea. 16 of the cholera cases occurred 2 weeks or longer after the second dose of vaccine (14 placebo recipients, 2 vaccinees). We also detected 14 symptomless infections (11 [7 placebo recipients, 4 vaccinees]) 2 weeks or longer after the second dose. The vaccine had significant protective efficacy against cholera (86% [95% Cl 37-97], p<0·01) but not against symptomless infection (42% [-96 to 85]). All cholera cases were in people of blood group O, who made up 76% of the study population (p<0·01). Two doses of WC/rBS vaccine, given 1 to 2 weeks apart, provide rapid, short-term protection against symptomatic cholera in adult South Americans, who are predominantly of blood group O. Long-term efficacy studies in Peruvian adults and children are under way.
Division of Preventive Medicine, Walter Reed Army Institute of Research, Washington DC, USA (J L Sanchez MD); United States Navy Medical Research Institute Detachment, Lima, Peru (B Vasquez MD, R E Begue MD, R Meza BS, G Castellares MD, C Cabezas MD, D M Watts PhD); Department of Medical Microbiology and Immunology, University of Göteborg, Sweden (A-M Svennerholm MD); and Division of Communicable Diseases and Immunology, Walter Reed Army Institute of Research, Washington, DC, USA (J C Sadoff MD, D N Taylor MD)
Correspondence to: Dr Jose L Sanchez, US Army Medical Research Unit-Brazil, American Consulate-Rio, Unit 3501, APO AA 34030, Rio de Janeiro, Brazil
Introduction The cholera epidemic in Latin (Central and South) America has had enormous medical, economic, and social consequences. In 3 years nearly a million cases and more than 8000 deaths from cholera have been reported.’ In Peru, an estimated 495 million dollars were lost in 1991 in the areas of tourism, fishing, exports, and health.2 Although many lives have been saved with oral rehydration, many patients with severe cholera require initial intravenous rehydration. The cost of intravenous fluids is prohibitively expensive in the developing world. Improvement of the potable water supply and sanitation in Latin America is recognised as an important priority for the long-term reduction of enteric diseases.3 However, the necessary improvements in the infrastructure will require billions of dollars and decades to complete.’I The recent cholera epidemic has renewed interest in the development of safe and effective orally administered cholera vaccines. An oral cholera vaccine consisting of killed cholera organisms plus the B subunit of cholera toxin was well tolerated and provided significant protection against cholera and enterotoxigenic Escherichia coli (ETEC) in Bangladesh.4-7 This vaccine provided 85% efficacy against cholera in the first 6 months and 50% cumulative efficacy over 3 years. Although the Bangladesh trial was designed to test a regimen of three doses given with 6-week intervals, subgroup analyses suggested that a two-dose regimen was equally effective.6 A less expensive formulation of this vaccine has been developed with a recombinant form of the B subunit (rBS).8 This formulation (WC/rBS) was found to be safe and immunogenic in North American and Swedish adults and Peruvians of all ages.9-11 In the Bangladesh trial the vaccine had significantly lower than average efficacy among children aged 2-5 years (presumably because of a lack of previous immunity), among people infected with the El Tor biotype6 and among people of blood group 0.12 These factors could lower the efficacy of this vaccine in Latin America, since the outbreak was caused entirely by the El Tor biotype, the population had no previous immunity to cholera, and about 75% of people are of blood group 0. Because of these factors the WC/rBS vaccines had to be studied in controlled clinical trials before it could be recommended for widespread use.
Patients and methods Each dose of WC/rBS vaccine (SBL Vaccin AB, Stockholm, Sweden) consists of 1 mg recombinantly produced cholera toxin B subunit and 2-5X10"’ of each of the following Vibrio cholerae
1273
in 3 mL phosphate-buffered saline: heatkilled classical Inaba (strain Cairo 48), heat-killed classical Ogawa (strain Cairo 50), formalin-treated El Tor Inaba (strain Phil 6973), and formalin-treated classical Ogawa (strain Cairo 50). Each 3 mL dose of vaccine was administered with a sodium bicarbonate buffer (Samarin, Cederroths Nordic AB, Upplands Vasby, Sweden) in 150 mL water containing 3-8 g sodium bicarbonate, 2-4 g tartaric acid, and 0-81 g citric acid. The placebo consisted of a suspension of heat-inactivated E coli K12 strain (SBL Vaccin AB) in a concentration that matched the turbidity and appearance of the vaccine preparation. The agents were dispensed in 300 mL bottles and stored at 4-8°C. Each bottle was identified with a unique number; vaccine and placebo
organisms suspended
preparations were pre-coded. Volunteers were enrolled in January and March, 1994, from three military training centres near Lima, Peru (CITEN, IMAP, and FAP). To be eligible, volunteers had to be 17-65 years old, to be at the centre for a minimum of 3 months, and to give written informed consent. No subject had received cholera vaccine previously. The study was approved by the Institutional Review Board of the Peruvian Navy Surgeon General, the Walter Reed Army Institute of Research Human Use Review Committee, and the Human Subjects Research Review Board, Office of the Army Surgeon General. The main objective of this study was to assess the efficacy of the vaccine against cholera. The only question of interest was whether the vaccine was more (and not less) protective than placebo. Therefore, the study was designed to detect a vaccine efficacy of 70% or more at p<0-025 (one-tailed) with a power of 0-8. On the basis of previous experience of cholera outbreaks at the participating military bases we assumed that the cumulative frequency of cholera would be 2-5% during the 3 months of basic training, therefore 457to 1165subjects per group would be required. Randomisation was done in blocks of 10 (5 to vaccine, 5 to placebo) to ensure equal study groups. Adverse events were not actively monitored in this study and no samples were taken for serology. Passive surveillance for diarrhoea was carried out at the medical clinics serving the training centres. Doctors from the Peruvian armed forces attended all diarrhoea cases on a 24 h basis. Subjects in the study were frequently reminded to report any episode of diarrhoea. For any patient with diarrhoea who was taking part in the study, the information contained on his study identification card was recorded at presentation. Within 48 h his identity was confirmed against a master list of study participants kept at the US Navy Medical Research Institute Detachment (NAMRID). Information on non-participants with diarrhoea was collected but not used for efficacy analyses. Clinical information was collected on standard data forms and a stool-impregnated rectal swab was collected and stored in Cary-Blair transport medium. Samples were transported to NAMRID and processed within 48 h of collection. Conventional techniques were used to isolate V cholerae 01 and to find out the biotype and serotype of each isolate.When outbreaks occurred, epidemiological information and rectal swabs were also obtained from symptom-free people in the recruit groups most affected. Such cross-sectional surveys were carried out four times (Feb 3 at CITEN, and Feb 8, March 29, and April 6 at IMAP). The incidence of cholera in the study group had exceeded 2% by June 1, 1994. Since no cases of cholera had occurred in the bases for 6 weeks, we decided to end the study on June 15, 1994. Data files were verified and edited, including determination of case and vaccine status, before the study code was broken.The cumulative time-in-study (in person-weeks) of those receiving two correct doses of each agent was recorded and used as the denominator to calculate incidence rates for each study group. Subjects lost to follow-up after the second dose were assumed to contribute half of the period to the denominator in the analysis. According to our protocol, vaccination was defined as the ingestion of two full doses of vaccine or placebo given 7-14 days apart. A case of cholera was defined as non-bloody diarrhoea (three or more loose stools in a 24 h period) with onset 2 or more weeks after the second dose, associated with the isolation of
1274
RR=rate ratio, weeks.
PE=protective efficacy.
Incidence rate ?
cases
per 1000 person-
*p<0’01 (one-tailed) for comparison of vaccine versus placebo Table : Protective efficacy of vaccine V cholerae 01 1 from a faecal sample. Severe cholera was defined as diarrhoea of any duration plus two other symptoms or objective signs of dehydration (lethargy, severe thirst, rapid and feeble pulse, tachypnoea, decreased skin turgor, sunken eyes, low urine output).ABO blood group tests were done by the Lee-Vincent 15 method on fingerstick samples. Vaccine protective efficacy was estimated as: rotectlve e Protective efficacy= lcacy=
Attack rate in unvaccinated-attack Attack rate m unvaccinated
rate m
vaccmated
X 100
95% CI on protective efficacy were calculated by accepted methods for field trials of vaccines. Incidence rates, rate ratios (incidence rate in vaccinated/incidence rate in unvaccinated), and 95% CI on rate ratios were also calculated. Proportions were compared by the X2 or Fisher’s test (two-tailed).
Results The
study population consisted of 1563 Peruvian men (military recruits and students) with an average age of 19 years (range 16-45). The subjects were randomly allocated vaccine (781) or placebo (782). 1426 (91%) received two doses of vaccine or placebo. Surveillance continued for 12 596 person-weeks among vaccinees and 12 794 person-weeks for placebo recipients. Losses to follow-up were minimal and similar in the two groups (22 [3%] of 710 vaccine recipients and 16 [2%] of 716 placebo recipients). The two-dose vaccination regimen was given within 2 weeks of the recruits’ arrival on-base. Of the 1426 subjects, 97% (1379) received two doses with an interval of 7-11 days between doses and 3% (47) received the two doses 27 days apart. Between Jan 4 and June 15, 1994, 383 (25%) of the study subjects went to the clinic with diarrhoea and samples were taken. Cases started to occur 2-3 weeks after the second dose was given at the CITEN site (late January, 1994). Two outbreaks occurred at the IMAP site within 2-4 weeks of vaccination (early February and early April, 1994). These two study sites accounted for all but 6 of the diarrhoea cases reported among study subjects. V cholerae 01was isolated from 38 (10%) of diarrhoea cases among study participants. The overall attack rate for cholera in the study population was therefore 2-4% (38 cases among 1563 participants). A further 20 cholera cases were identified among non-participants (estimated population 6000). V cholerae El Tor Ogawa was isolated from 54 subjects and V cholerae El Tor Inaba from 4. Of the 38 cases among study participants, 16 (42%) satisfied the criteria for inclusion in the efficacy analysis (two-dose recipients, 2 or more weeks after the second dose). Of the 22 cases not included in the efficacy analysis, 21 occurred before the second dose was given and only 1 case (in a placebo recipient) occurred shortly after the second dose. In addition, 14 symptomless infections were diagnosed among study subjects (all caused by V cholerae El Tor Ogawa). 11 (79%) of these occurred among two-dose
recipients.
after the second dose when the interval between doses is 7-14 days. Thus, protection can be provided within 3 weeks of the first dose. This short interval could make the vaccine a useful public health intervention for the prevention of cholera epidemics. We also found that WC/rBS cholera vaccine was very effective in protecting Peruvian adults against El Tor cholera. The protection achieved in our study was similar to that shown by Clemens et al4 in the first 6 months of a large-scale efficacy study in Bangladesh (77% [95% CI 25-92]). Protection was achieved after only two doses given in a short time. A regimen of this nature is much simpler to implement than one with two or three doses given 6 weeks apart. These data support observations in Swedish adults that equivalent immunogenicity can be attained when two doses of this vaccine are given with intervals of 1-6 weeks. 17 It also supports previous findings from Bangladesh that two doses were as effective as three in preventing cholera after 3 years of surveillance.6 This study is the first demonstration that the vaccine is effective in Latin America, where El Tor is the only biotype and where the population is predominantly of blood group 0. In Bangladesh’2 there was a significantly lower protective efficacy among subjects of blood group 0 than among other subjects (40 vs 76%). Others have noted the predisposition for severe cholera among hospital Ínpatients2,18.19 as well as in volunteers2O of blood group 0. Our study supports those observations. The cholera attack rates (2-3%) seen during the military outbreaks are about ten times higher than those being experienced among civilian populations in Peru.2>2’ The epidemic setting observed in this trial is similar to that observed in other military forces and refugee populations and to conditions at the onset of the cholera epidemic in Peru in 1991. It is well known that the epidemiological features of epidemic cholera differ from those where cholera has become endemic.22 The pattern of outbreaks with high attack rates among adults seen in epidemic cholera, when contaminated food items and water are often implicated in the transmission of V cholerae, more closely resembles the conditions in our
shortly
Figure: Cumuiative frequency of cholera among study groups p
2 of the 16 cases of cholera in two-dose recipients occurred among vaccinees and 14 among placebo recipients (table). The WC/rBS vaccine therefore conferred significant protection against symptomatic cholera (protective efficacy 86% [95% CI 37-97%]). Protection was evident 2 weeks after the second dose and was maintained for the rest of the study period (figure) There were 4 symptomless cholera infections in the vaccine group compared with 7 among placebo recipients. The protective efficacy against symptomless infection was not significant (42% [-96 to 83%]). Rates of diarrhoeal illness (non-cholera diarrhoea) 2 weeks or longer after the second dose were similar in vaccine and placebo groups. Similarly, there was no difference in diarrhoea incidence during the 7-13 days after the second dose. Further investigation of cholera infections among 780 placebo recipients was done to assess potential associations between blood group 0 and infection status, as well as severity of infection. 32 V cholerae 01 infections occurred among placebo recipients (infection rate 4-1%). In 22 of these cases, the patient had symptoms (cholera attack rate 2-8%). Symptomatic cholera occurred only in individuals of blood group 0 (22 of 558 with blood group 0 vs 0 of 180 of other blood types, p<0-01). Only 2 of the 22 cases were severe, and both were in people of blood group 0. For symptomless infections the rate of V cholerae 01infection was similar in 0 and non-0 blood groups (8 of 558 [1’4%] vs 2 of 180 [1-1%], respectively). Because very few cholera cases were detected in our study, no subgroup efficacy analyses could be done for blood group or severity of illness.
study. The short time between vaccination and the onset of cholera may have provided optimum conditions for the assessment of the protective efficacy of WC/rBS vaccine. Before the vaccine can be recommended for use on a wider scale, large-scale efficacy trials among choleraendemic populations in Lima and Arequipa must be completed. Studies of vaccine effectiveness and efficiency will also be required before this vaccine can be recommended for use in a public health programme.23
Discussion Since cholera appeared in Peru in 1991 there have been annual outbreaks of diarrhoea (and cholera) at military recruit training bases during the warm summer months (unpublished). These outbreaks have occurred soon after recruits arrive on-base, and confirmed cholera-attack rates are as high as 16% among affected units. This year we were able to vaccinate incoming recruits before outbreaks occurred. We found that if the second dose of vaccine could be given 2 weeks before a cholera outbreak
began, Only 1
References 1 2
3 4 5
1988; 158: 60-68.
it could case
weeks of the
provide significant protection (figure). occurred, in a placebo recipient, within 2 second dose. Immunity seems to develop
Pan American Health Organization. Cholera situation in the Americas. Epidemiol Bull 1994; 15: 13-16. Gotuzzo E, Cieza J, Estremadoyro L, Seas C. Cholera: lessons from the epidemic in Peru. Infect Dis Clin N Am 1994; 8: 183-205. Pan American Health Organization. Cholera vaccine evaluation. Epidemiol Bull 1991; 12: 5-10. Clemens JD, Sack DA, Harris JR, et al. Field trial of oral cholera vaccines in Bangladesh. Lancet 1986; ii: 124-27. Clemens JD, Harris JR, Sack DA, et al. Field trial of oral cholera vaccines in Bangladesh: results of one year of follow-up. J Infect Dis
6
Clemens JD, Sack DA, Harris JR, et al. Field trial of oral cholera vaccines in Bangladesh: results from three-year follow-up. Lancet 1990; 339: 270-73.
1275
7
Clemens JD, Stanton BF, Chakraborty J, et al. B subunit-whole cell and whole cell-only oral vaccines against cholera: studies on reactogenicity and immunogenicity. J Infect Dis 1987; 155: 79-85.
8
Holmgren J, Clemens J, Sack DA, Svennerholm A-M.
New cholera
vaccines. Vaccine 1989; 7: 94-96. Sanchez JL, Trofa A, Taylor DN, et al. Safety and immunogenicity of the oral, inactivated, whole cell plus recombinant B subunit of cholera toxin (WC/rBS) cholera vaccine in North American volunteers. J Infect Dis 1993; 167: 1446-49. 10 Jertborn M, Svennerholm A-M, Holmgren J. Safety and immunogenicity of an oral recombinant cholera B subunit-whole cell vaccine in Swedish volunteers. Vaccine 1992; 10: 130-32. 11 Begue RE, Castellares G, Ruiz R, et al. Community-based assessment of safety and immunogenicity of the whole cell plus recombinant B subunit (WC/rBS) oral cholera vaccine in Peru. Vaccine
9
(in press).
14
Kelly MT, Hickman-Brenner FW, Farmer III JJ. Vibrio. In: Balows A, Hausler WJ, Herrmann KL, Isenberg HD, Shadomy HJ, eds. Manual of clinical microbiology, 5th edition. Washington DC: American Society for Microbiology, 1991: 384-95. Bennish ML. Cholera: pathophysiology, clinical features, and treatment.
In: Wachsmuth IK, Blake PA, Olsvik O, eds. Vibrio cholerae
Society for Microbiology,
Garraty G. Blood group antigens and antibodies and pretransfusion compatibility testing. In: Rose NR, Conway de Macario E, Fahey JL, Friedman H, Penn GM, eds. Manual of clinical laboratory immunology, 4th edition. Washington DC: American Society for Microbiology, 1992: 308-19. 16 Orenstein WA, Bernier RH, Dondero TJ, et al. Field evaluation of vaccine efficacy. Bull WHO 1985; 63: 1055-68. 17 Jertborn M, Svennerholm A-M, Holmgren J. Evaluation of different
15
immunization schedules for oral cholera B subunit-whole cell vaccine in Swedish volunteers. Vaccine 1993; 11: 1007-12.
D, Paquio AS. ABO blood groups and cholera. Ann Hum Biol 1977; 4: 489-92. Chaudhuri A. Cholera and blood groups. Lancet 1977; ii: 404.
18 Barua 19
12 Clemens JD, Sack DA, Harris JR, et al. ABO blood groups and cholera: new observations on specificity of risk and modification of vaccine efficacy. J Infect Dis 1989; 159: 770-73.
13
and cholera. Washington DC: American 1994: 229-55.
20 Levine
MM, Nalin DR, Rennels MB, et al. Genetic cholera. Ann Hum Biol 1979; 6: 359-74.
21
susceptibility
to
Castellares G, Hayashi KE, et al. Diarrheal disease in Peru after the introduction of cholera. Am J Trop Med Hyg (in press).
Begue RE,
22 Glass 23
RI, Black RE. The epidemiology of cholera. In: Barua D, Greenough III WB, eds. Cholera. New York: Plenum, 1992: 129-54. Willems J, Sanders C. Cost-effectiveness and cost-benefit analyses of vaccines. J Infect Dis 1981; 144: 4486-93.
Teamwork
For
long time, general practice was epitomised by a partnership between patient and personal doctor, who worked single-handed. In the past few decades, however, medical care has become ever more complex and one consequence has been the emergence of primary-care teams.! Two characteristics make primary care unique in comparison with other sorts of medical practice-general practitioners provide "first" medical care for all categories of patients; and they provide care that is continuous over time. Potential members of the team should share, at least in essence, these approaches to patients. This paper reviews the development of the primary-care team from the perspective of the general practitioner-the partners in the team, the motives behind the formation of partnerships, and the consequences for individual care of patients. The
a
reference, which is
an important feature of a discipline; administrative staff usually work under supervision. In primary medical care, however, administrative and logistic support have become so important for day-to-day patient care that "protoprofessionalisation" is under way-a switch from administrative functions to direct involvement in patient care. 21 Various terms are used for this new function (practice nurse in the UK, practice assistant in the Netherlands), mirroring a mixed background of origin. Medical professionals in the community include occupational health officers and public health officers-for
whereas
primary-care team
Influence of health-service structure In health-care systems where there is a functional differentiation of tasks between primary and secondary care, most disciplines working in the community will have a primary-care orientation. Table 1 lists the professionals who offer unselected and ongoing care for broad categories of patients. The term discipline warrants some comment. The paramedical workers, psychologists, and social workers have an independent professional frame of
Department of General Practice and Social Medicine, University of Nijmegen, PO Box 9101,6500 HB Nijmegen, Netherlands (Prof C van Weel FRCGP)
1276
Table 1:
Primary care: the team