- Email: [email protected]
Coprologically diagnosing Anoplocephala perfoliata in the presence of A. magna
Accepted Manuscript Title: Coprologically diagnosing Anoplocephala perfoliata in the presence of A. magna Author: Alejandro Boh´orquez Ar´anzazu Meana N´elida F. Pato M´onica Luz´on PII: DOI: Reference:
S0304-4017(14)00232-5 http://dx.doi.org/doi:10.1016/j.vetpar.2014.04.023 VETPAR 7230
To appear in:
Veterinary Parasitology
Received date: Revised date: Accepted date:
20-12-2013 16-4-2014 21-4-2014
Please cite this article as: Boh´orquez, A., Meana, A., Pato, N.F., Luz´on, M.,Coprologically diagnosing Anoplocephala perfoliata in the presence of A. magna, Veterinary Parasitology (2014), http://dx.doi.org/10.1016/j.vetpar.2014.04.023 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
30
Alejandro Bohórquez1, Aránzazu Meana1, Nélida F. Pato2, Mónica Luzón1* 1
ip t
Departamento de Sanidad Animal, Facultad de Veterinaria, Universidad Complutense de Madrid, Avenida de Puerta de Hierro s/n, 28040 Madrid, España. 2 Hospital Clínico Veterinario, Facultad de Veterinaria, Universidad Alfonso X El Sabio, Avenida Universidad 1, Villanueva de la Cañada, 28691 Madrid, España.
cr
* Corresponding author. Tel.: +34 3943948; fax: +34 91 3943908. E-mail address: [email protected] (M. Luzón).
d
M
an
us
Abstract Current copro-diagnostic tests for Anoplocephala perfoliata show high variation in their sensitivity and given the morphological similarity of Anoplocephala spp. eggs, this could be related to the presence of Anoplocephala magna alone or co-existing with A. perfoliata. In the present study, coprology was significantly more sensitive (p < 0.01) at detecting A. magna than A. perfoliata. This difference was independent of the parasite burden and was greater when testing was limited to horses with mature or gravid tapeworms. A. magna infection was strongly linked to young horses (≤ 2 years). The eggs of A. magna are smaller. Using 15 μm and 70 µm cut-offs for oncosphere diameter and the major shell bisector length, respectively, the eggs of A. perfoliata were identified with 100% sensitivity, 97% specificity and 98% sensitivity, 84% specificity. The use of these two morphometric variables would therefore be useful for the coproidentification of A. perfoliata in countries where both species coexist. Keywords: Sensitivity; specificity; copro-diagnosis; Anoplocephala perfoliata; Anoplocephala magna; host-age distribution.
Introduction
te
29
Coprologically diagnosing Anoplocephala perfoliata in the presence of A. magna
Ac ce p
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Anoplocephala perfoliata (Cyclophyllidea, Anoplocephalidae) is a common
31
intestinal tapeworm that affects horses worldwide. Mounting evidence suggests that A.
32
perfoliata is responsible for several intestinal disorders (review by Gasser et al., 2005),
33
and its accurate detection is crucial for both clinical and epidemiological purposes.
34
Currently available diagnostic methods are based on the assumption that A. perfoliata is
35
the only epidemiologically significant tapeworm to parasitize domestic horses.
36
However, this assumption does not hold true in countries such as USA or Spain. In
37
USA, Anoplocephala magna, a cestode of limited pathogenicity, is at least as prevalent
38
as A. perfoliata, (Lichtenfels, 1975). In Spain, since the first detection of A. magna in
39
abattoirs in northern and central regions (Meana et al., 2002), this cestode has been
Page 1 of 18
2 40
frequently detected throughout the country, although with different seasonality from A.
41
perfoliata (Meana et al., 2005). To improve the diagnosis of A. perfoliata infection, several methods have been
43
developed including the detection of parasite eggs in faeces using improved
44
centrifugation and flotation techniques, though these methods have shown considerable
45
variation in their sensitivity: from 40 – 62% (Gasser et al., 2005) to 75% (Rebhein et
46
al., 2010). Besides differences inherent to each technique, some factors that could
47
contribute to this variability are the intensity of infection, the irregular shedding of
48
proglottids or the maturity state of the tapeworms, but the possibility of
49
infection/coinfection with A. magna, should also be considered.
an
us
cr
ip t
42
Thus, in areas where both these two Anoplocephala species coexist, the copro-
51
diagnosis of equine cestodoses has been hindered because of the morphological
52
similarity between the eggs of the two species. The information on shell diameter is
53
scarce and applicable to A. perfoliata (65-80 µm, see reviews by Soulsby, 1987;
54
Tarazona Vilas, 1999; Gasser et al., 2005) but not to A. magna. According to different
55
reviews, the eggs of A. magna are either smaller (50 – 60 µm in Soulsby, 1987) or
56
larger (70 – 80 µm in Tarazona Vilas, 1999) than those of A. perfoliata. The limited
57
data on the oncosphere indicates a larger diameter for A. perfoliata (16 µm) than A.
58
magna (8 µm) (Tarazona Vilas, 1999).
d
te
Ac ce p
59
M
50
The differences in pathogeny and ecology between both tapeworm species
60
justify the necessity of differential diagnostic methods. In order to evaluate the
61
reliability and accuracy of the copro-diagnosis in the detection of A. perfoliata in the
62
presence of A. magna, the sensitivity of the copro-diagnosis of A. perfoliata and A.
63
magna was examined in horses identified by necropsy as positive for Anoplocephala
64
spp. as single or double infections. The distribution of each species according to the age
Page 2 of 18
3 65
and sex of the host was also examined. Finally, the reliability of two morphometric
66
variables (major shell bisector length and oncosphere diameter) for the identification of
67
eggs of both Anoplocephala species was assessed.
Materials and Methods
70
1. Sensitivity of coprology as a diagnostic test
cr
69
ip t
68
The digestive tracts of horses slaughtered in Spain were dissected and 130,
72
classified as positive for Anoplocephala spp. infection by visual inspection, used as the
73
reference standard. All cestodes recovered from each digestive tract were fixed in 70%
74
ethanol for their identification and recording of tapeworm burdens. Faeces were taken
75
directly from the rectum of each infected animal and 40 g analyzed by the
76
sedimentation/flotation method described by Meana et al. (1998) to detect
77
Anoplocephala spp. eggs. The sensitivity of coprology as a diagnostic test was then
78
calculated for each Anoplocephala species.
te
d
M
an
us
71
The stage of infection of each horse was also established according to
80
morphological criteria. For A. perfoliata, tapeworms were classified as “immature”
81
(lancet-shaped) versus “mature” (triangular-shaped) according to Schuster (1991). For
82
A. magna, small (< 9 cm long) uniformly shaped tapeworms were defined as
83
“immature” as compared to longer “mature” tapeworms with wider, mostly darker,
84
terminal segments. The sensitivity of coprology as a diagnostic test was also calculated
85
for each species using the horses with mature tapeworms as the reference standard.
Ac ce p
79
86
The presence of eggs in terminal proglottids of tapeworms classified as “mature”
87
was investigated in a group of horses with a low parasite burden (< 30 tapeworms). The
88
sensitivity of coprology was also calculated for each species using the horses with
Page 3 of 18
4 89
gravid tapeworms as the reference standard. Statistical analyses of sensitivities were
90
performed by the Chi squared test.
91 2. Host distribution of species
ip t
92
Of the 130 animals selected as positive for Anoplocephala spp., only 94 animals
94
could be identified by sex (45 males and 49 females) and age (66 young, i.e. ≤ 2-years-
95
old, and 28 adults, i.e. >2-years-old) as the slaughterhouses records were not always
96
accessible. Of these, animals with a single infection (n = 43 A. perfoliata and n = 29 A.
97
magna) were selected to establish the distribution of species according to host age or
98
sex.
an
us
cr
93
100
M
99 3. Egg morphology
Anoplocephala eggs were collected from about 10 specimens of each
102
Anoplocephala spp., in both cases coming from more than two animals, in order to
103
increase the quality of the samples representation. The last gravid segment of each
104
tapeworm was sectioned across the rear transverse line with a scalpel, and the eggs
105
collected by shaking the sectioned segments inside Eppendorf® tubes (volume 1.5 ml)
106
filled with tap water. For each species, the contents collected from different tapeworms
107
were mixed together.
te
Ac ce p
108
d
101
Under a light microscope (Nikon), 100 eggs of each species were examined at a
109
magnification of x400 to obtain two measurements: a) the diameter of the oncosphere,
110
and b) the length of the major shell bisector (Figure 1). The sensitivity (Se) and
111
specificity (Sp) of each morphometric character to identify each Anoplocephala species
112
were calculated by selecting an optimal cut-off for each measurement based on the ROC
113
(Receiver Operating Characteristic) curve (Noordhuizen et al., 1997). Briefly, the ROC
Page 4 of 18
5 curve displays the relationship between the true positive fraction (Se) and the false
115
positive fraction (1-Sp) over the range of possible cut-off values. In the ROC analysis,
116
the optimal cut-off value is that with “sensitivity plus specificity” (Se + Sp) at a
117
maximum. The area under the ROC curve (AUC) is a measurement of the validity of
118
the analysis, with a value of 1 representing the perfect analysis. These calculations were
119
carried out using the “Threshold value of a diagnostic test” option of Win Episcope 2.0.
cr
ip t
114
Results
122
1. Sensitivity of coprology as a diagnostic test
an
121
us
120
Tapeworm burdens, infecting parasites and coprological results are provided in
124
Table 1. Sixty four animals harboured A. perfoliata and 35 animals harboured A. magna
125
as a single infection, while 35 animals harboured both Anoplocephala spp. infections.
126
Tapeworm burdens were mostly low (1-30 tapeworms) for both species (68% A.
127
perfoliata and 89% A. magna). High A. perfoliata burdens (> 100 tapeworms) were in
128
the ranges 115 – 396 in single infections and 138 – 1,113 in double infections, whereas
129
A. magna burdens were 130 – 170 in double infections. The diagnostic sensitivities of
130
coprology were 11% and 51% respectively for A. perfoliata and A. magna in single
131
infections and 58% in double infections. Considering single and double infections
132
together, sensitivities were 26% for A. perfoliata and 55% for A. magna. Statistical
133
analysis indicated a significantly higher (p<0.01) sensitivity to detect A. magna than A.
134
perfoliata when comparing both single and single plus double infections. For both
135
species, low tapeworm burdens predominated over moderate (31 – 100 tapeworms) or
136
high burdens in animals returning a positive coprology test.
Ac ce p
te
d
M
123
137
As shown in Table 2, in the subset of horses with mature tapeworms, sensitivity
138
was significantly lower (p<0.01) in animals parasitized with A. perfoliata (15%) than
Page 5 of 18
6 139
with A. magna (75%). Similar results were obtained in animals harbouring gravid
140
tapeworms (20% vs 79%, p < 0.01).
141 2. Host distribution of species
ip t
142
The presence of A. magna as a single infection (n = 29) was significantly
144
associated (p < 0.01) with host age (90% young, n = 26 vs 10% adults, n = 3). These
145
results differed significantly (p < 0.01) from those recorded for single A. perfoliata
146
infections (n = 43) (53% foals, n = 23 vs 47% adults, n = 20) in which no age
147
association was observed. Only three adult females harboured A. magna as a single
148
infection; of them, one was a pregnant mare at the time of necropsy and other mare had
149
foaled six months earlier. No host sex predilection was observed of either species (A.
150
perfoliata: 40% males, n = 17 vs 60% females, n = 26; A magna: 52% males, n = 15 vs
151
48% females, n = 14).
d
M
an
us
cr
143
Of the subset of animals showing A. magna as a double infection (22 animals
153
that were excluded from the host distribution analysis), five were adults (3 females and
154
2 males).
Ac ce p
155
te
152
156
3. Identification according to egg morphology
157
3.a. Oncosphere diameter (Figure 1, a)
158
As shown in Table 3, the oncosphere diameter of the A. perfoliata eggs
159
examined ranged from 15 to 25 µm (mean ± SD 17.78 ± 1.28), with a mode of 17.5 µm
160
(84% of A. perfoliata eggs). For A. magna eggs, oncosphere diameters were 10 to 17.5
161
µm (mean ± SD 12.21 ± 1.34), with a mode of 12.5 µm (44% of A. magna eggs). In
162
100% of A. perfoliata eggs, oncospheres were ≥ 15 µm whereas in 97% of A. magna
163
eggs they were < 15 µm. According to the ROC analysis, the optimal threshold to
Page 6 of 18
7 distinguish between A. perfoliata and A. magna eggs was a diameter of 15 µm. Using
165
this cut-off, eggs of A. perfoliata would be distinguished from A. magna with a
166
sensitivity of 100% and specificity of 97% (Figure 2). The AUC was 0.99 (95% C.I.:
167
0.98 – 0.99), indicating high accuracy.
ip t
164
168 3.b. Major shell bisector length (Figure 1, b)
cr
169
As shown in Table 3, the A. perfoliata egg shell’s major bisector was 65 – 112.5
171
µm long (mean ± SD 80.63 ± 8.14) with values in the range 77.5 – 85 µm showing
172
highest frequencies (58% of the eggs). Lengths for A. magna ranged from 57.5 to 75
173
µm (mean ± SD 65.06 ± 3.94), and the highest frequency range was 62.5 – 67.5 µm
174
(52% of the eggs). In 98% of A. perfoliata eggs, shell bisectors were ≥ 70 µm, whereas
175
in 84% of A. magna eggs they were < 70 µm. According to the ROC analysis, 70 µm
176
was selected as the optimal cut-off bisector length to distinguish between the eggs of A.
177
perfoliata and A. magna with a sensitivity of 98% and specificity of 84% (Figure 3).
178
The AUC was 0.97 (95% C.I.: 0.96 – 0.98), indicating high accuracy.
180 181
an
M
d
te
Ac ce p
179
us
170
Discussion
This study reveals that as a diagnostic test, coprology is significantly more
182
sensitive to detect A. magna than A. perfoliata. This difference was independent of the
183
tapeworm burden since low burdens of both Anoplocephala spp. were predominant in
184
animals testing positive by coprology. The higher sensitivity of the test observed here
185
for A. magna over A. perfoliata was even more evident in horses parasitized with
186
mature or gravid tapeworms. Using this sample, any possible biases due to an immature
187
tapeworm stage were ruled out. The most likely explanation for it is the larger size of A.
188
magna, whose long strobila is made up of wide segments that break-off in groups when
Page 7 of 18
8 gravid. In contrast, A. perfoliata is much smaller and its gravid proglottids form and
190
break away at a much slower rate. Another finding of this study was the preference
191
shown by A. magna for young (≤ 2 years) horses. Thus, 90% of single infections by A.
192
magna were detected in young animals. Only three adult females were parasitized by A.
193
magna as a single infection, one of them being pregnant and other a postparturient
194
mare. Tapeworm infections have been related to host age in other Anoplocephalidae
195
infections such as those of cattle with Moniezia spp. (Ramajo Martín and Muro Álvarez,
196
1999). Although more work is needed, the finding of A. magna infection in young
197
animals and pregnant/postparturient mares could be related to factors associated to
198
young age or pregnancy.
an
us
cr
ip t
189
The higher coprological sensitivity detected towards A. magna and the strong
200
preference shown by this species for young animals mean that past epidemiological
201
studies have probably overestimated the prevalence of A. perfoliata in areas where both
202
occur, especially those performed in young horses, given the morphological similarity
203
of Anoplocephala spp. eggs.
d
te
Despite the morphological similarity of the eggs, the morphometric analysis
Ac ce p
204
M
199
205
revealed differences in the oncosphere diameter and major shell bisector length. In both
206
cases, measurements were larger for A. perfoliata than A. magna. The difference in the
207
major shell bisector is in line with the description by Soulsby (1987), yet is inconsistent
208
with the findings of Tarazona Vilas (1999) who reported the similar or greater size of A.
209
magna eggs. As for the oncosphere diameter, the figures are consistent with data
210
provided by Tarazona Vilas (1999), although this author’s oncosphere measurements
211
for A. magna were smaller.
212
According to those morphometric data, both cut-offs, oncosphere diameter and
213
major shell bisector length, were assigned to distinguish the eggs of A. perfoliata from
Page 8 of 18
9 those of A. magna with high accuracy, especially with the oncosphere diameter cut-off.
215
The diagnostic implication of this finding is remarkable as no clear morphometric
216
differentiation had been reported so far. Control programs with an epidemiological basis
217
must rely on specific diagnoses. In this respect, current immunodiagnostic methods may
218
not discriminate between both Anoplocephala spp. (Bohórquez et al., 2012) and
219
possibly the identification will rely on molecular tools. For the moment, the use of these
220
two morphometric traits could help the coprological identification of A. perfoliata.
221
Despite its many limitations, in most laboratories coprology is still a routine technique
222
of diagnosing a parasite disease. Notwithstanding the demonstrated sensitivity problems
223
of the copro-diagnosis of A. perfoliata, we propose that this method would be especially
224
useful in young animals reared in countries like the USA or Spain where the
225
coexistence of both species has been demonstrated (Lichtenfels, 1975; Meana et al.,
226
2002, 2005).
d
M
an
us
cr
ip t
214
229 230 231 232 233 234
Conflict of interest statement
The authors declare that no competing interest exist.
Ac ce p
228
te
227
Acknowledgements
This work was supported by Virbac, S.A. The authors thank M. Martínez-
Izquierdo for help with the microscopy measurements.
235
References
236
Bohórquez, A., Meana, A., Luzón, M. 2012. Differential diagnosis of equine cestodosis
237
based on E/S and somatic Anoplocephala perfoliata and Anoplocephala magna
238
antigens. Vet. Parasitol. 190, 87-94.
Page 9 of 18
10 239
Gasser, R.B., Williamson, R.M.C., Beveridge, I., 2005. Anoplocephala perfoliata of horses
240
– significant scope for further research, improved diagnosis and control.
241
Parasitology 131, 1-13.
243
Lichtenfels, J.R., 1975. Helminths of domestic equids. Proceedings of the
ip t
242
Helminthological Society of Washington, 42 (special issue). 92 pp.
Meana, A., Luzón, M., Corchero, J., Gómez-Bautista, M., 1998. Reliability of
245
coprological diagnosis of Anoplocephala perfoliata infection. Vet. Parasitol. 74,
246
79-83.
248
us
Meana, A., Mateos, A., Pato, N.F., Pérez, J., 2002. First report of Anoplocephala magna
an
247
cr
244
(Abildgaard, 1789) in Spain. Rev. Ibér. Parasitol. 3-4, 93-95. Meana, A., Pato, N.F., Martín, R., Mateos, A., Pérez-García, J., Luzón, M., 2005.
250
Epidemiological studies on equine cestodes in central Spain: Infection pattern
251
and population dynamics. Vet. Par. 30, 233-240.
d
M
249
Noordhuizen, J. P. T. M., Frankena, K., van der Hood, C.M., Graat, E.A.M., 1997.
253
Application of quantitative methods in veterinary epidemiology. Wageningen
254
Pers, Wageningen, The Netherlands. Pp. 445.
Ac ce p
te
252
255
Ramajo Martín, V., Muro Álvarez, A., 1999. Parasitosis de los rumiantes. Cestodosis
256
digestivas. In: Cordero del Campillo, M., Rojo Vázquez, F.A. (Eds.),
257
Parasitología Veterinaria. McGraw-Hill Interamericana, Madrid: 229-234.
258
Rebhein, S., Lindner, T., Visser, M., Winter, R., 2010. Evaluation of a double
259
centrifugation technique for the detection of Anoplocephla eggs in horse faeces.
260
J Helminthol. 1-6. Schuster, R., 1991. [Morphometric analysis of an
261
Anoplocephala perfoliata population]. Angewandte Parasitologie 32, 105-111.
262
[In German].
Page 10 of 18
11 263
Soulsby, E. J. L., 1987. Parasitología y enfermedades parasitarias en los animales
264
domésticos (1st ed. in Spanish, Translation of the 7th ed. in English). Ed.
265
Interamericana, Mexico. 823 pp. Tarazona Vilas, J. M., 1999. Parasitosis de los équidos. Cestodosis intestinales:
267
anoplocephalidosis. In: Cordero del Campillo, M., Rojo Vázquez, F.A. (Eds.),
268
Parasitología Veterinaria. McGraw-Hill Interamericana, Madrid: 540-545.
cr
ip t
266
Ac ce p
te
d
M
an
us
269
Page 11 of 18
12 269
Table 1. Tapeworm burdens and coprological results of 130 horses parasited by A.
270
perfoliata and A. magna as a single or double infection.
A.perfoliata
TAPEWORM BURDEN 1-30
31-100
> 100
TOTAL ANIMALS N=130
65 (17)+
17 (6) +
13 (2) +
95
COPROLOGICAL RESULTS FAECES + 25
SENSITIVITY
ip t
INFECTION
26%
single
double
59 (29) +
32 (15) +
27 (14) +
5 (1) +
4 (4) +
3 (3) +
3 (3) +
1 (1) +
31
(n)+: animals testing positive by coprology
272
( )*: 95% Confidence Interval
3 (3) +
66
35
18
11%
(5 – 21)* 58% (40 – 75)*
36
55% (42 – 66)*
18
51% (34 – 69)*
31
18
58% (40 – 75)*
Ac ce p
271
273
0
7
us
7 (5) +
64
an
19 (12) +
8 (1) +
M
A. magna
10 (1) +
d
double
46 (5) +
te
single
cr
(18 – 36)*
Page 12 of 18
13 Table 2. Coprological results of 81 horses with mature tapeworms of A. perfoliata (n =
274
53) or A. magna (n = 28), as a single or double infection, and of 24 horses with gravid
275
tapeworms of A. perfoliata (n = 10) or A. magna (n = 14), as a single or double
276
infection. COPROLOGICAL RESULTS
NUMBER
FAECES
SENSITIVITY
ANIMALS
+
(95% C.I.)
Mature A.perfoliata
53
8
15% (7 – 27)
single
48
7
double*
5
1
Mature A. magna
28
21
single
25
18
double*
3
3
Gravid A.perfoliata
10
2
single
10
double*
0
Gravid A. magna
14
single
5
M
an
75% (55 – 89)
20% (3 – 52)
2
d
-
te
double*
cr
OF
us
INFECTION
ip t
273
9
11
79% (52 – 94)
4 7
(*): Animals with double Anoplocephala spp. infection but having only mature or gravid
278
tapeworms of the mentioned species.
279
280
281
282
Ac ce p
277
283 284 285
Page 13 of 18
14 Table 3. Results of measuring the oncosphere diameter (left) and the major shell
286
bisector (right) in 100 eggs of A. magna and 100 eggs of A. perfoliata. The black line
287
represents the optimal cut-off between A. perfoliata and A. magna values according to
288
the ROC analysis. Grey area below the black line represents false positive values. Green
289
area above the black line represents false negative values.
292
0
10
57.5
11.25
0
28
58.75
12.5
0
44
13.75
0
15
15
4
1
17.5
84
2
20
10
0
22.5
1
25
1
Number of eggs
A. perfoliata
A. magna 2
0
3
60
0
5
61.25
0
7
62.5
0
18
63.75
0
15
65
2
12
66.25
0
10
0
67.5
0
11
70
10
8
72
5
4
75
10
5
77.5
18
0
80
7
0
82.5
19
0
85
14
0
87.5
7
0
90
2
0
92.5
2
0
100
1
0
105
1
0
112.5
2
0
M
an
0
d
291
10
te
0
Ac ce p
290
bisector (µm)
A. perfoliata A. magna
cr
diameter (µm)
Major shell
Number of eggs
us
Oncosphere
ip t
285
Page 14 of 18
15 Figure captions
293
Figure 1. Egg of Anoplocephala spp., showing the oncosfere diameter (a) and the major
294
shell bisector (b).
295
Figure 2. Sensitivity (Se) and specificity (Sp) for each potential cut-off value
296
oncosphere diameter to distinguish between A. perfoliata and A. magna eggs, obtained
297
from the observation of 100 eggs of each species.
298
Figure 3. Sensitivity (Se) and specificity (Sp) for each potential cut-off value major
299
shell bisector to distinguish between A. perfoliata and A. magna eggs, obtained from the
300
observation of 100 eggs of each species.
an
us
cr
ip t
292
Ac ce p
te
d
M
301
Page 15 of 18
i
Figure
b a
Ac
ce pt
ed
M an
us
cr
Figure1
Page 16 of 18
cr
i
Figure
M an
us
Figure 2 1 0.9 0.9 0.8 0.8
ed
0.7 0.7
Se
ce pt
0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2
Sp
Ac
Se or Sp
0.6 0.6
0.1 0.1
0
10
11.75
12.5
13.75
15
17.5
20
22.5
25
Cut-off values (µm)
Page 17 of 18
cr
i
Figure
M an
us
Figure 3
11 0.9 0.8
ed
0.7 0.7
0.4
0.3 0.2
0.1
ce pt
0.5
Se Sp
Ac
Se or Sp
0.6
00
Cut-off values (µm) Page 18 of 18
S0304-4017(14)00232-5 http://dx.doi.org/doi:10.1016/j.vetpar.2014.04.023 VETPAR 7230
To appear in:
Veterinary Parasitology
Received date: Revised date: Accepted date:
20-12-2013 16-4-2014 21-4-2014
Please cite this article as: Boh´orquez, A., Meana, A., Pato, N.F., Luz´on, M.,Coprologically diagnosing Anoplocephala perfoliata in the presence of A. magna, Veterinary Parasitology (2014), http://dx.doi.org/10.1016/j.vetpar.2014.04.023 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
30
Alejandro Bohórquez1, Aránzazu Meana1, Nélida F. Pato2, Mónica Luzón1* 1
ip t
Departamento de Sanidad Animal, Facultad de Veterinaria, Universidad Complutense de Madrid, Avenida de Puerta de Hierro s/n, 28040 Madrid, España. 2 Hospital Clínico Veterinario, Facultad de Veterinaria, Universidad Alfonso X El Sabio, Avenida Universidad 1, Villanueva de la Cañada, 28691 Madrid, España.
cr
* Corresponding author. Tel.: +34 3943948; fax: +34 91 3943908. E-mail address: [email protected] (M. Luzón).
d
M
an
us
Abstract Current copro-diagnostic tests for Anoplocephala perfoliata show high variation in their sensitivity and given the morphological similarity of Anoplocephala spp. eggs, this could be related to the presence of Anoplocephala magna alone or co-existing with A. perfoliata. In the present study, coprology was significantly more sensitive (p < 0.01) at detecting A. magna than A. perfoliata. This difference was independent of the parasite burden and was greater when testing was limited to horses with mature or gravid tapeworms. A. magna infection was strongly linked to young horses (≤ 2 years). The eggs of A. magna are smaller. Using 15 μm and 70 µm cut-offs for oncosphere diameter and the major shell bisector length, respectively, the eggs of A. perfoliata were identified with 100% sensitivity, 97% specificity and 98% sensitivity, 84% specificity. The use of these two morphometric variables would therefore be useful for the coproidentification of A. perfoliata in countries where both species coexist. Keywords: Sensitivity; specificity; copro-diagnosis; Anoplocephala perfoliata; Anoplocephala magna; host-age distribution.
Introduction
te
29
Coprologically diagnosing Anoplocephala perfoliata in the presence of A. magna
Ac ce p
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Anoplocephala perfoliata (Cyclophyllidea, Anoplocephalidae) is a common
31
intestinal tapeworm that affects horses worldwide. Mounting evidence suggests that A.
32
perfoliata is responsible for several intestinal disorders (review by Gasser et al., 2005),
33
and its accurate detection is crucial for both clinical and epidemiological purposes.
34
Currently available diagnostic methods are based on the assumption that A. perfoliata is
35
the only epidemiologically significant tapeworm to parasitize domestic horses.
36
However, this assumption does not hold true in countries such as USA or Spain. In
37
USA, Anoplocephala magna, a cestode of limited pathogenicity, is at least as prevalent
38
as A. perfoliata, (Lichtenfels, 1975). In Spain, since the first detection of A. magna in
39
abattoirs in northern and central regions (Meana et al., 2002), this cestode has been
Page 1 of 18
2 40
frequently detected throughout the country, although with different seasonality from A.
41
perfoliata (Meana et al., 2005). To improve the diagnosis of A. perfoliata infection, several methods have been
43
developed including the detection of parasite eggs in faeces using improved
44
centrifugation and flotation techniques, though these methods have shown considerable
45
variation in their sensitivity: from 40 – 62% (Gasser et al., 2005) to 75% (Rebhein et
46
al., 2010). Besides differences inherent to each technique, some factors that could
47
contribute to this variability are the intensity of infection, the irregular shedding of
48
proglottids or the maturity state of the tapeworms, but the possibility of
49
infection/coinfection with A. magna, should also be considered.
an
us
cr
ip t
42
Thus, in areas where both these two Anoplocephala species coexist, the copro-
51
diagnosis of equine cestodoses has been hindered because of the morphological
52
similarity between the eggs of the two species. The information on shell diameter is
53
scarce and applicable to A. perfoliata (65-80 µm, see reviews by Soulsby, 1987;
54
Tarazona Vilas, 1999; Gasser et al., 2005) but not to A. magna. According to different
55
reviews, the eggs of A. magna are either smaller (50 – 60 µm in Soulsby, 1987) or
56
larger (70 – 80 µm in Tarazona Vilas, 1999) than those of A. perfoliata. The limited
57
data on the oncosphere indicates a larger diameter for A. perfoliata (16 µm) than A.
58
magna (8 µm) (Tarazona Vilas, 1999).
d
te
Ac ce p
59
M
50
The differences in pathogeny and ecology between both tapeworm species
60
justify the necessity of differential diagnostic methods. In order to evaluate the
61
reliability and accuracy of the copro-diagnosis in the detection of A. perfoliata in the
62
presence of A. magna, the sensitivity of the copro-diagnosis of A. perfoliata and A.
63
magna was examined in horses identified by necropsy as positive for Anoplocephala
64
spp. as single or double infections. The distribution of each species according to the age
Page 2 of 18
3 65
and sex of the host was also examined. Finally, the reliability of two morphometric
66
variables (major shell bisector length and oncosphere diameter) for the identification of
67
eggs of both Anoplocephala species was assessed.
Materials and Methods
70
1. Sensitivity of coprology as a diagnostic test
cr
69
ip t
68
The digestive tracts of horses slaughtered in Spain were dissected and 130,
72
classified as positive for Anoplocephala spp. infection by visual inspection, used as the
73
reference standard. All cestodes recovered from each digestive tract were fixed in 70%
74
ethanol for their identification and recording of tapeworm burdens. Faeces were taken
75
directly from the rectum of each infected animal and 40 g analyzed by the
76
sedimentation/flotation method described by Meana et al. (1998) to detect
77
Anoplocephala spp. eggs. The sensitivity of coprology as a diagnostic test was then
78
calculated for each Anoplocephala species.
te
d
M
an
us
71
The stage of infection of each horse was also established according to
80
morphological criteria. For A. perfoliata, tapeworms were classified as “immature”
81
(lancet-shaped) versus “mature” (triangular-shaped) according to Schuster (1991). For
82
A. magna, small (< 9 cm long) uniformly shaped tapeworms were defined as
83
“immature” as compared to longer “mature” tapeworms with wider, mostly darker,
84
terminal segments. The sensitivity of coprology as a diagnostic test was also calculated
85
for each species using the horses with mature tapeworms as the reference standard.
Ac ce p
79
86
The presence of eggs in terminal proglottids of tapeworms classified as “mature”
87
was investigated in a group of horses with a low parasite burden (< 30 tapeworms). The
88
sensitivity of coprology was also calculated for each species using the horses with
Page 3 of 18
4 89
gravid tapeworms as the reference standard. Statistical analyses of sensitivities were
90
performed by the Chi squared test.
91 2. Host distribution of species
ip t
92
Of the 130 animals selected as positive for Anoplocephala spp., only 94 animals
94
could be identified by sex (45 males and 49 females) and age (66 young, i.e. ≤ 2-years-
95
old, and 28 adults, i.e. >2-years-old) as the slaughterhouses records were not always
96
accessible. Of these, animals with a single infection (n = 43 A. perfoliata and n = 29 A.
97
magna) were selected to establish the distribution of species according to host age or
98
sex.
an
us
cr
93
100
M
99 3. Egg morphology
Anoplocephala eggs were collected from about 10 specimens of each
102
Anoplocephala spp., in both cases coming from more than two animals, in order to
103
increase the quality of the samples representation. The last gravid segment of each
104
tapeworm was sectioned across the rear transverse line with a scalpel, and the eggs
105
collected by shaking the sectioned segments inside Eppendorf® tubes (volume 1.5 ml)
106
filled with tap water. For each species, the contents collected from different tapeworms
107
were mixed together.
te
Ac ce p
108
d
101
Under a light microscope (Nikon), 100 eggs of each species were examined at a
109
magnification of x400 to obtain two measurements: a) the diameter of the oncosphere,
110
and b) the length of the major shell bisector (Figure 1). The sensitivity (Se) and
111
specificity (Sp) of each morphometric character to identify each Anoplocephala species
112
were calculated by selecting an optimal cut-off for each measurement based on the ROC
113
(Receiver Operating Characteristic) curve (Noordhuizen et al., 1997). Briefly, the ROC
Page 4 of 18
5 curve displays the relationship between the true positive fraction (Se) and the false
115
positive fraction (1-Sp) over the range of possible cut-off values. In the ROC analysis,
116
the optimal cut-off value is that with “sensitivity plus specificity” (Se + Sp) at a
117
maximum. The area under the ROC curve (AUC) is a measurement of the validity of
118
the analysis, with a value of 1 representing the perfect analysis. These calculations were
119
carried out using the “Threshold value of a diagnostic test” option of Win Episcope 2.0.
cr
ip t
114
Results
122
1. Sensitivity of coprology as a diagnostic test
an
121
us
120
Tapeworm burdens, infecting parasites and coprological results are provided in
124
Table 1. Sixty four animals harboured A. perfoliata and 35 animals harboured A. magna
125
as a single infection, while 35 animals harboured both Anoplocephala spp. infections.
126
Tapeworm burdens were mostly low (1-30 tapeworms) for both species (68% A.
127
perfoliata and 89% A. magna). High A. perfoliata burdens (> 100 tapeworms) were in
128
the ranges 115 – 396 in single infections and 138 – 1,113 in double infections, whereas
129
A. magna burdens were 130 – 170 in double infections. The diagnostic sensitivities of
130
coprology were 11% and 51% respectively for A. perfoliata and A. magna in single
131
infections and 58% in double infections. Considering single and double infections
132
together, sensitivities were 26% for A. perfoliata and 55% for A. magna. Statistical
133
analysis indicated a significantly higher (p<0.01) sensitivity to detect A. magna than A.
134
perfoliata when comparing both single and single plus double infections. For both
135
species, low tapeworm burdens predominated over moderate (31 – 100 tapeworms) or
136
high burdens in animals returning a positive coprology test.
Ac ce p
te
d
M
123
137
As shown in Table 2, in the subset of horses with mature tapeworms, sensitivity
138
was significantly lower (p<0.01) in animals parasitized with A. perfoliata (15%) than
Page 5 of 18
6 139
with A. magna (75%). Similar results were obtained in animals harbouring gravid
140
tapeworms (20% vs 79%, p < 0.01).
141 2. Host distribution of species
ip t
142
The presence of A. magna as a single infection (n = 29) was significantly
144
associated (p < 0.01) with host age (90% young, n = 26 vs 10% adults, n = 3). These
145
results differed significantly (p < 0.01) from those recorded for single A. perfoliata
146
infections (n = 43) (53% foals, n = 23 vs 47% adults, n = 20) in which no age
147
association was observed. Only three adult females harboured A. magna as a single
148
infection; of them, one was a pregnant mare at the time of necropsy and other mare had
149
foaled six months earlier. No host sex predilection was observed of either species (A.
150
perfoliata: 40% males, n = 17 vs 60% females, n = 26; A magna: 52% males, n = 15 vs
151
48% females, n = 14).
d
M
an
us
cr
143
Of the subset of animals showing A. magna as a double infection (22 animals
153
that were excluded from the host distribution analysis), five were adults (3 females and
154
2 males).
Ac ce p
155
te
152
156
3. Identification according to egg morphology
157
3.a. Oncosphere diameter (Figure 1, a)
158
As shown in Table 3, the oncosphere diameter of the A. perfoliata eggs
159
examined ranged from 15 to 25 µm (mean ± SD 17.78 ± 1.28), with a mode of 17.5 µm
160
(84% of A. perfoliata eggs). For A. magna eggs, oncosphere diameters were 10 to 17.5
161
µm (mean ± SD 12.21 ± 1.34), with a mode of 12.5 µm (44% of A. magna eggs). In
162
100% of A. perfoliata eggs, oncospheres were ≥ 15 µm whereas in 97% of A. magna
163
eggs they were < 15 µm. According to the ROC analysis, the optimal threshold to
Page 6 of 18
7 distinguish between A. perfoliata and A. magna eggs was a diameter of 15 µm. Using
165
this cut-off, eggs of A. perfoliata would be distinguished from A. magna with a
166
sensitivity of 100% and specificity of 97% (Figure 2). The AUC was 0.99 (95% C.I.:
167
0.98 – 0.99), indicating high accuracy.
ip t
164
168 3.b. Major shell bisector length (Figure 1, b)
cr
169
As shown in Table 3, the A. perfoliata egg shell’s major bisector was 65 – 112.5
171
µm long (mean ± SD 80.63 ± 8.14) with values in the range 77.5 – 85 µm showing
172
highest frequencies (58% of the eggs). Lengths for A. magna ranged from 57.5 to 75
173
µm (mean ± SD 65.06 ± 3.94), and the highest frequency range was 62.5 – 67.5 µm
174
(52% of the eggs). In 98% of A. perfoliata eggs, shell bisectors were ≥ 70 µm, whereas
175
in 84% of A. magna eggs they were < 70 µm. According to the ROC analysis, 70 µm
176
was selected as the optimal cut-off bisector length to distinguish between the eggs of A.
177
perfoliata and A. magna with a sensitivity of 98% and specificity of 84% (Figure 3).
178
The AUC was 0.97 (95% C.I.: 0.96 – 0.98), indicating high accuracy.
180 181
an
M
d
te
Ac ce p
179
us
170
Discussion
This study reveals that as a diagnostic test, coprology is significantly more
182
sensitive to detect A. magna than A. perfoliata. This difference was independent of the
183
tapeworm burden since low burdens of both Anoplocephala spp. were predominant in
184
animals testing positive by coprology. The higher sensitivity of the test observed here
185
for A. magna over A. perfoliata was even more evident in horses parasitized with
186
mature or gravid tapeworms. Using this sample, any possible biases due to an immature
187
tapeworm stage were ruled out. The most likely explanation for it is the larger size of A.
188
magna, whose long strobila is made up of wide segments that break-off in groups when
Page 7 of 18
8 gravid. In contrast, A. perfoliata is much smaller and its gravid proglottids form and
190
break away at a much slower rate. Another finding of this study was the preference
191
shown by A. magna for young (≤ 2 years) horses. Thus, 90% of single infections by A.
192
magna were detected in young animals. Only three adult females were parasitized by A.
193
magna as a single infection, one of them being pregnant and other a postparturient
194
mare. Tapeworm infections have been related to host age in other Anoplocephalidae
195
infections such as those of cattle with Moniezia spp. (Ramajo Martín and Muro Álvarez,
196
1999). Although more work is needed, the finding of A. magna infection in young
197
animals and pregnant/postparturient mares could be related to factors associated to
198
young age or pregnancy.
an
us
cr
ip t
189
The higher coprological sensitivity detected towards A. magna and the strong
200
preference shown by this species for young animals mean that past epidemiological
201
studies have probably overestimated the prevalence of A. perfoliata in areas where both
202
occur, especially those performed in young horses, given the morphological similarity
203
of Anoplocephala spp. eggs.
d
te
Despite the morphological similarity of the eggs, the morphometric analysis
Ac ce p
204
M
199
205
revealed differences in the oncosphere diameter and major shell bisector length. In both
206
cases, measurements were larger for A. perfoliata than A. magna. The difference in the
207
major shell bisector is in line with the description by Soulsby (1987), yet is inconsistent
208
with the findings of Tarazona Vilas (1999) who reported the similar or greater size of A.
209
magna eggs. As for the oncosphere diameter, the figures are consistent with data
210
provided by Tarazona Vilas (1999), although this author’s oncosphere measurements
211
for A. magna were smaller.
212
According to those morphometric data, both cut-offs, oncosphere diameter and
213
major shell bisector length, were assigned to distinguish the eggs of A. perfoliata from
Page 8 of 18
9 those of A. magna with high accuracy, especially with the oncosphere diameter cut-off.
215
The diagnostic implication of this finding is remarkable as no clear morphometric
216
differentiation had been reported so far. Control programs with an epidemiological basis
217
must rely on specific diagnoses. In this respect, current immunodiagnostic methods may
218
not discriminate between both Anoplocephala spp. (Bohórquez et al., 2012) and
219
possibly the identification will rely on molecular tools. For the moment, the use of these
220
two morphometric traits could help the coprological identification of A. perfoliata.
221
Despite its many limitations, in most laboratories coprology is still a routine technique
222
of diagnosing a parasite disease. Notwithstanding the demonstrated sensitivity problems
223
of the copro-diagnosis of A. perfoliata, we propose that this method would be especially
224
useful in young animals reared in countries like the USA or Spain where the
225
coexistence of both species has been demonstrated (Lichtenfels, 1975; Meana et al.,
226
2002, 2005).
d
M
an
us
cr
ip t
214
229 230 231 232 233 234
Conflict of interest statement
The authors declare that no competing interest exist.
Ac ce p
228
te
227
Acknowledgements
This work was supported by Virbac, S.A. The authors thank M. Martínez-
Izquierdo for help with the microscopy measurements.
235
References
236
Bohórquez, A., Meana, A., Luzón, M. 2012. Differential diagnosis of equine cestodosis
237
based on E/S and somatic Anoplocephala perfoliata and Anoplocephala magna
238
antigens. Vet. Parasitol. 190, 87-94.
Page 9 of 18
10 239
Gasser, R.B., Williamson, R.M.C., Beveridge, I., 2005. Anoplocephala perfoliata of horses
240
– significant scope for further research, improved diagnosis and control.
241
Parasitology 131, 1-13.
243
Lichtenfels, J.R., 1975. Helminths of domestic equids. Proceedings of the
ip t
242
Helminthological Society of Washington, 42 (special issue). 92 pp.
Meana, A., Luzón, M., Corchero, J., Gómez-Bautista, M., 1998. Reliability of
245
coprological diagnosis of Anoplocephala perfoliata infection. Vet. Parasitol. 74,
246
79-83.
248
us
Meana, A., Mateos, A., Pato, N.F., Pérez, J., 2002. First report of Anoplocephala magna
an
247
cr
244
(Abildgaard, 1789) in Spain. Rev. Ibér. Parasitol. 3-4, 93-95. Meana, A., Pato, N.F., Martín, R., Mateos, A., Pérez-García, J., Luzón, M., 2005.
250
Epidemiological studies on equine cestodes in central Spain: Infection pattern
251
and population dynamics. Vet. Par. 30, 233-240.
d
M
249
Noordhuizen, J. P. T. M., Frankena, K., van der Hood, C.M., Graat, E.A.M., 1997.
253
Application of quantitative methods in veterinary epidemiology. Wageningen
254
Pers, Wageningen, The Netherlands. Pp. 445.
Ac ce p
te
252
255
Ramajo Martín, V., Muro Álvarez, A., 1999. Parasitosis de los rumiantes. Cestodosis
256
digestivas. In: Cordero del Campillo, M., Rojo Vázquez, F.A. (Eds.),
257
Parasitología Veterinaria. McGraw-Hill Interamericana, Madrid: 229-234.
258
Rebhein, S., Lindner, T., Visser, M., Winter, R., 2010. Evaluation of a double
259
centrifugation technique for the detection of Anoplocephla eggs in horse faeces.
260
J Helminthol. 1-6. Schuster, R., 1991. [Morphometric analysis of an
261
Anoplocephala perfoliata population]. Angewandte Parasitologie 32, 105-111.
262
[In German].
Page 10 of 18
11 263
Soulsby, E. J. L., 1987. Parasitología y enfermedades parasitarias en los animales
264
domésticos (1st ed. in Spanish, Translation of the 7th ed. in English). Ed.
265
Interamericana, Mexico. 823 pp. Tarazona Vilas, J. M., 1999. Parasitosis de los équidos. Cestodosis intestinales:
267
anoplocephalidosis. In: Cordero del Campillo, M., Rojo Vázquez, F.A. (Eds.),
268
Parasitología Veterinaria. McGraw-Hill Interamericana, Madrid: 540-545.
cr
ip t
266
Ac ce p
te
d
M
an
us
269
Page 11 of 18
12 269
Table 1. Tapeworm burdens and coprological results of 130 horses parasited by A.
270
perfoliata and A. magna as a single or double infection.
A.perfoliata
TAPEWORM BURDEN 1-30
31-100
> 100
TOTAL ANIMALS N=130
65 (17)+
17 (6) +
13 (2) +
95
COPROLOGICAL RESULTS FAECES + 25
SENSITIVITY
ip t
INFECTION
26%
single
double
59 (29) +
32 (15) +
27 (14) +
5 (1) +
4 (4) +
3 (3) +
3 (3) +
1 (1) +
31
(n)+: animals testing positive by coprology
272
( )*: 95% Confidence Interval
3 (3) +
66
35
18
11%
(5 – 21)* 58% (40 – 75)*
36
55% (42 – 66)*
18
51% (34 – 69)*
31
18
58% (40 – 75)*
Ac ce p
271
273
0
7
us
7 (5) +
64
an
19 (12) +
8 (1) +
M
A. magna
10 (1) +
d
double
46 (5) +
te
single
cr
(18 – 36)*
Page 12 of 18
13 Table 2. Coprological results of 81 horses with mature tapeworms of A. perfoliata (n =
274
53) or A. magna (n = 28), as a single or double infection, and of 24 horses with gravid
275
tapeworms of A. perfoliata (n = 10) or A. magna (n = 14), as a single or double
276
infection. COPROLOGICAL RESULTS
NUMBER
FAECES
SENSITIVITY
ANIMALS
+
(95% C.I.)
Mature A.perfoliata
53
8
15% (7 – 27)
single
48
7
double*
5
1
Mature A. magna
28
21
single
25
18
double*
3
3
Gravid A.perfoliata
10
2
single
10
double*
0
Gravid A. magna
14
single
5
M
an
75% (55 – 89)
20% (3 – 52)
2
d
-
te
double*
cr
OF
us
INFECTION
ip t
273
9
11
79% (52 – 94)
4 7
(*): Animals with double Anoplocephala spp. infection but having only mature or gravid
278
tapeworms of the mentioned species.
279
280
281
282
Ac ce p
277
283 284 285
Page 13 of 18
14 Table 3. Results of measuring the oncosphere diameter (left) and the major shell
286
bisector (right) in 100 eggs of A. magna and 100 eggs of A. perfoliata. The black line
287
represents the optimal cut-off between A. perfoliata and A. magna values according to
288
the ROC analysis. Grey area below the black line represents false positive values. Green
289
area above the black line represents false negative values.
292
0
10
57.5
11.25
0
28
58.75
12.5
0
44
13.75
0
15
15
4
1
17.5
84
2
20
10
0
22.5
1
25
1
Number of eggs
A. perfoliata
A. magna 2
0
3
60
0
5
61.25
0
7
62.5
0
18
63.75
0
15
65
2
12
66.25
0
10
0
67.5
0
11
70
10
8
72
5
4
75
10
5
77.5
18
0
80
7
0
82.5
19
0
85
14
0
87.5
7
0
90
2
0
92.5
2
0
100
1
0
105
1
0
112.5
2
0
M
an
0
d
291
10
te
0
Ac ce p
290
bisector (µm)
A. perfoliata A. magna
cr
diameter (µm)
Major shell
Number of eggs
us
Oncosphere
ip t
285
Page 14 of 18
15 Figure captions
293
Figure 1. Egg of Anoplocephala spp., showing the oncosfere diameter (a) and the major
294
shell bisector (b).
295
Figure 2. Sensitivity (Se) and specificity (Sp) for each potential cut-off value
296
oncosphere diameter to distinguish between A. perfoliata and A. magna eggs, obtained
297
from the observation of 100 eggs of each species.
298
Figure 3. Sensitivity (Se) and specificity (Sp) for each potential cut-off value major
299
shell bisector to distinguish between A. perfoliata and A. magna eggs, obtained from the
300
observation of 100 eggs of each species.
an
us
cr
ip t
292
Ac ce p
te
d
M
301
Page 15 of 18
i
Figure
b a
Ac
ce pt
ed
M an
us
cr
Figure1
Page 16 of 18
cr
i
Figure
M an
us
Figure 2 1 0.9 0.9 0.8 0.8
ed
0.7 0.7
Se
ce pt
0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2
Sp
Ac
Se or Sp
0.6 0.6
0.1 0.1
0
10
11.75
12.5
13.75
15
17.5
20
22.5
25
Cut-off values (µm)
Page 17 of 18
cr
i
Figure
M an
us
Figure 3
11 0.9 0.8
ed
0.7 0.7
0.4
0.3 0.2
0.1
ce pt
0.5
Se Sp
Ac
Se or Sp
0.6
00
Cut-off values (µm) Page 18 of 18