- Email: [email protected]
Changes in ocular onchocerciasis after two rounds of community-based ivermectin treatment in a holo-endemic onchocerciasis focus
TRANSACTIONS OFTHEROYALSCEIETY OFTROPICAL MEDICINE ANDHYGIENE (1991)
Changes in ocular onchocerciasis after ivermectin treatment in a holo-endemic K. Y. Dadzie, J. Remme and G. De Sole Ouagadougou, Burkina Faso
In a longitudinal study to determine the effect of annual community-based treatment of ocular onchocerciasis with ivermectin, the population living in the 3 most affected villages in the holo-endemic onchocerciasis focus of Asubende in Ghana were re-examined 16 and 24 months after initiating treatment . Ocular microfilarial loads had decreasedto very low levels in nearly all of the 334 examined persons who were treated twice. Only very few subjects had ocular loads of 32 microfilariae or more in the anterior chamber of the eye, but this was not associatedwith deterioration of ocular lesions. Important regression of both early and advanced lesions of the anterior segment of the eye was observed, which was highly statistically significant with respect to iridocyclitis. Lesions of the posterior segment of the eye remained stable. Though no systematic change in the visual acuity of the population was observed, 3 new casesof blindness occurred in persons who already had eye lesions at such. an advanced stage that ivermectin treatment could no longer affect the outcome, The results suggest that annual ivermectin treatment is adequate to control onchocercal ocular diseaseeven in populations with very high endemicity levels. Introduction
Ivermectin has been shown to be a safe microfilaritide suitable for large scale treatment of onchocerciasis (DE SOLEet d., 1989a, 1990; REMME ef al., 1990). However its full potential as a tool for the long-term control of onchocercal morbidity and the prevention of blindness is not yet known (BROWNet al.. 1990). In order to determine the effect of annual community-based ivermectin treatment on the eye, longitudinal studies were started in a population living in a holo-endemic onchocerciasis focus in Ghana. The first results, 4 and 12 months after the first treatment round, have been reported elsewhere (DADZIE et al., 1990b). In this paper, changes in ocular onchocerciasis 16 and 24 months after initiating annual community-based ivermectin treatment are reported.
The study population was drawn from the 3 most endemic communities in the isolated savanna onchocerciasis focus in the Asubende area in Ghana. where a community trial of ivermectin was started in 1987 (DE SOLEet al., 1989b; REMMEet al., 1989a). In these communities the prevalence of microfilariae in skin snips ranged from 85.6% to 87*3%, and the community microfilarial load from 58 to 73 microfilariae per skin snip. Of the total census population of 864 persons, 586 aged 5 years and over were examined
267
two rounds of community-based onchocerciasis focus
Onchocerciasis Control Programme in West Africa,
Abstract
Patients and Methods Study population
85, 267-271
B.P.
549,
ophthahnologically before the first mass treatment (DADZIE et al., 199Ob),and 581 and 586 persons were examined at the 16 months and 24 months surveys respectively. Drug administration
In October 1987 and October 1988 ivermectin tablets were given as a single oral dose of approximately 150 @kg body weight to all those in the study area who were eligible, under the direct supervision of a medical officer or a nurse. The exclusion criteria used during the second treatment were the same as durin the first treatment round, extent for women breast! eedina infants who were now excluded when breastfeeding &r infant of up to the age of 3 months instead of 1 month. The manufacturer of the drug had effected this change. Ophtkalnwlogical
examination
Detailed ophthalmological examination of all subjects aged 5 years and above was carried out before treatment and the examination was repeated 4,2 1, 16 and 24 months thereafter. The test of visual acuity was conducted using Sjiigren’s illiterate hand testtype in broad daylight. Slit lamp examination of the anterior segment of the eye, as well as fundus examination with the direct and indirect onhthalmoscope after pupil dilatation, was carried out in the onhthalmic mobile clinic followinn the onchocerciasis control project (OCP) protocol asieported previously (DADZIE et al., 1990a).All ophthalmological examinations were carried out by the same examiner and no reference was made to the record forms of previous examinations in the course of the survey. Classification of ocular lesions.
The 4 typical onchocercal eye lesions, viz. sclerosing keratitis, iridocyclitis, optic atrophy and choroido-retinitis. were classified into 3 erades: no lesion (absence of signs of any lesion), &rly lesion, and advanced lesion. A detailed description of the classification criteria has been given in the previous report (DADZIE et al., 1990a). Statistical analysis
Data processing was done using routine OCP computer proarams and SPSS/PCfor the final analvsis. 437 paher& had had a complete ophthalmological examination during the pre-treatment survey and those at 16 and 24 months after the start of the treatment programme. Of those patients, 334 had been treated twice with ivermectin with a yearly interval, 51 patients had been treated during the first treatment round only, 38 patients during the second round only, and 14 patients had been excluded from both treatments becauseof the exclusion criteria given
268 above. Only data from the right eye of the 14 non-treated patients and 334 patients treated twice were considered for the analysis. The trends in mean microfilarial loads during the full 2-year follow-up period could be shown only for the 289 patients who were treated twice and examined at all 5 surveys. Geometric mean microfilarial loads were calculated using the log (x+1) transformation. Differences in microfilarial loads were tested for statistical significance using the Wilcoxon signed rank sum test for comparisons between surveys. The 4 typical onchocercal eye lesions were allocated a score of 0 for no lesion, 1 for early lesion and 2 for advanced lesion. The significance of changes in ocular lesions within the treated and non-treated groups were tested using the Wilcoxon signed rank sum test on pre- and post-treatment scores for each lesion. For the comparison of the changes between the two groups the Mann-Whitney U test was applied to the difference between pre- and post-treatment scoresin caseswith a change in score. In 20 persons it was not possible to arrive at a classification for all lesions as the relevant signs had been obscured, usually by some other onchocercal pathology. In 18 subjects the view of the posterior segment was completely obstructed during at least one examination, and in 9 of them during all 3 examinations. For similar reasons it was not possible to obtain counts of microfilariae in the anterior chamber of the eye for 7 persons. These caseswith missing data were excluded from the analysis for the respective lesion or microfilarial load. 110 100
Results The Figure shows the trends in the mean number of microfilariae in the anterior chamber of the eye (MFAC) and in the mean number of dead microfilariae in the cornea (DMFC) of the 289 persons treated twice who were present at all 5 examinations. The mean microfilarial loads fell to low levels 4 months after each treatment round but showed a statistically significant increase 12 months thereafter (P
1st treatment
t
90 80 70 80 50 40 30 20 10 0 0
4
8
12
16
20
24
Number of months after first treatment Figure. Relative changes in ocular microfikarial loads after two rounds of ivermectin treatment; results for 289 patients; DMFC=mean dead microfihwiie in the cornea, MFAC=mean number of microfilariae in the anterior chamber.
number of
269 sion of anterior segment lesions. Table 2 shows the changes in distribution of onchocercal lesions of the anterior segment of the eye in patients treated twice. The classification of the presenceand severity of each lesion 16 and 24 months after the start of treatment is shown in relation to the pre-treatment classification of the sameeye in order to allow the assessmentof both random variation and systematic changes. At 16 months, 17 classified initially as having no sclerosing keratitis were reclassified as having an early lesion, but this was balanced by 19 cases with early sclerosing keratitis who were re-classified as having no lesion. Such results may be explained by random variation in the diagnosis of borderline lesions. The findings for advanced sclerosing keratitis are different. At 16 months only one new advanced lesion, which was initially classified as an early lesion, was reported, compared to 9 caseswith advanced lesions which had regressed, 6 to early lesions and 3 to no lesion. At 24 months, 14 of 19 cases with advanced sclerosing keratitis were re-classified as early or no lesion, and these changesin the treated group were of borderline statistical significance (P=O%).
The changesin the distribution of iridocyclitis were even more striking, and highly statistically significant. Of 36 caseswith early iridocyclitis pre-treatment, 32 were classified as negative after 24 months while not a single casemoved from negative to early lesion. Four caseswith advanced lesions were classified as negative for iridocyclitis at 24 months. With regard to choroido-retinitis, changes from no lesion to early lesioq, and from early lesion to advanced lesion, and vice versa, were similar and can be interpreted as random varmtion (Table 3). The same was true for optic atrophy. Table 3 shows that the pattern of ocular diseaseof the posterior segment of the eye remained stable during the lirst 2 years after the start of ivermectin treatment and that there was no statistically significant change during that period. The visual acuities of the 334 persons treated twice did not show any systematic change during the study period. However, 3 people in the study population had gone blind during the 24 months after the start of the community-based treatment, 1 after 12 months and the other 2 after 16 months. Two were men, aged 42 and 51 years respectively, who presented at the first examination with secondary glaucoma. Although
Table 1. Post-treatment distribution of microlihrial loads in the anterior chamber of the eye (MFAC) of 327 persons who were treated twice in relation to pre-treatment loads (percentage per pre-treatment group ia parentheses)
MFAC pre-treatment
No. of MFAC 16 months after star;;f trea3yTnt MFAC 24 months after start of treatment Persons 0 l40 l4lO32
0
144
l-
42
4-
31
lO-
78
32-
32
Total
327
$0)
$0,
(OVO)
(lf.0)
139 (96.5)
(2!8)
(017)
cove,
$0,
(9?2)
$4)
$0,
(2:)
$0)
(St;)
(9!5)
(2!4)
(2:)
(i-0,
(Io?O)
(,90,
(,90,
$0)
(090,
(7%)
(9.37)
(3!2)
$0)
(65.4)
(22.6) (O!O)
$3)
(6148s) (2?6)
(6!4)
(5!1)
(lf3)
(15!6)
(12:5)
(25!0)
(l:“-b,
(:!3)
$3)
(1~0)
(8;?7)
(S:!O) (1:*6) 301 (92.0) $4)
$5, (t8)
(9’4)
$5)
(f.2)
(k)
(3&) 256 (78.3)
$1)
$5) $5)
Table 2. Classification of onchocercal lesions of tbe anterior segment of the eye 16 and 24 months after the start of annual ivermectin mass treatment in relation to the pre-treatment classification
Type of lesion 16-monthfollow-up Sclerosing keratitis
Iridocyclitis 24-month follow-up
Sclerosing keratitis
Iridocyclitis
Post-treatment lesions Pre-treatment Non-treatedgroup Patients treated twice lesion None Early AdvancedNone Early Advanced Significanceof change None Early
10
2
-
;
i
; -
I 2
Advanced None Early Advanced
11
None Early Advanced None
11 10 1’:
1
-
7
;
-
-
2
Early Advanced
-
255 19 3 274 31 5
17 20 6
-1 10
-i -
; 14
253 22 3 274 32 4
19 16 11
-i 5
-i 2
13
-2
Within non-treated group f’ri*;T Within treated group Between the two groups P=O*34 Within non-treated group M&my Within treated group
Within non-treatedgroup P=l*O
Within treated group Between the two groups Within non-treated group Within treated group Between the two groups
P=OGl P=O*67 P=l-0 P
270 Table 3. Classilication of onchocercal ivermectin mass treatment in relation
Type of lesion Choroido-retinitis Optic atrophy
lesions of the posterior segment of the eye 24 months after the start of annual to the pre-treatment classification
Post-treatment lesions Pre-treatment Non-treatment group Patients treated twice None Early Advanced None Early Advanced Significance of change lesion None 10 269 6 Within non-treated group : I 6 7 -i Within treated group Early Advanced 6 13 Between the two groups 9 265 Within non-treated group None Early 1 ; ! 10 1; -4 Within treated group Advanced 1 3 16 Betweenthe two groups
they were treated symptomatically, the inevitable end of blindness could not be averted. The third was a 22 year old woman with advanced post-neuritic optic atrophy, for whom there was likewise no remedy available. The 2 men had been treated twice with ivermectin but the woman only once, as she had been excluded from the second treatment because of pregnancy. On the other hand, a 38 year old woman, who, because of corneal oedema resulting from bilateral secondary glaucoma could not count fingers at 3 m with either eye and who was by definition blind at the first examination, had responded to treatment with oral acetazolamideand topical steroids and mydriatics and was no longer classified as blind at the last surveys. She has also been treated twice with ivermectin .
Discussion
In order to determine the long-term impact of ivermectin treatment in the control of on&ocercal ocular disease. longitudinal oDhthahnoloeica1 followup examinations are needed td document carefully the changes in ocular disease occurring in communities included in mass treatment programmes (BROWN et al., 1990). Since the risk of developing onchocercal eye lesions is directly related to the intensity of infection, reflected by the microfilarial density in the eye and- in the skin (ANDERSONet al.; 1974; THYLEFORS& BRINKMANN. 1977: REMME et al.. 1989b), it is particularly imp&ant to do such studied in communities with high endemicity levels. In this respect the present study is unique as it reports the ocular changes after annual community-based treatment with ivermectin in a population with exceptionally high loads of ocular microfilariae. Ivermectin treatment resulted in a major reduction in ocular microfilarial loads 4 months after each treatment round. However, during the subsequent 8 months there was a sit&icant increase in the number of microfilariae in jhe eye, a phenomenon not observed in clinical trials (TAYLOR er al.. 1986: DADZIE et al., 1987, 1990a; ~VEWLANDet al:, 19881 VINGTAIN et al., 1988) but which is consistent with the post-treatment increase in skin microfilarial loads in the samepatients (REMME et al., 1990). Nevertheless, in most patients the ocular microfilarial loads remained very low, particularly after the second treatment, which resulted in a further reduction in the mean microfilarial loads in the eye. An exception to the above trend was observed in a few patients who, in spite of 2 ivermectin treatments, still had intermittently high loads of microfilariae in the anterior chamber of the eye. In none of them was
~$~ P=O.30 P=l.O P=O.34
P=O.78
worsening of the ocular diseaserecorded during the study period, while in half of them an improvement was observed. However, the number of casesinvolved was small and the period of observation relatively short. A longer period of observation, therefore, is necessaryto confirm whether intermittent high loads of ocular microfilariae in onchocerciasis patients under ivermectin treatment are unlikely to induce the development of eye lesions or cause the deterioration of existing ones. The previous report showed that early lesions of the anterior segment of the eyes tended to regress after the first round of ivermectin mass treatment and that the regression was particularly significant 4 months after treatment (DADZIE et al.. 1990a1.The nresent study not only cdnfirmed this oGervati& but px&ded further evidence that the regression of lesions is not only limited to early lesions of the anterior segment of the eye but involves advanced lesions also. The control group in this study, consisting of 14 subjects who had never been treated becauseof the exclusion criteria, was very small and this prevents a meaningful comparison with the group treated twice. Nevertheless, no tendency to regression among the non-treated patients was observed. The regression of iridocyclitis in treated patients was highly significant and, after 2 treatments, early iridocyclitis was hardly seen. Furthermore, as was noted in the previous report also (DADZIE et al., 1990b), there were several patients with iridocyclitis at the advanced stage of development who had regressedto negative when the synechiae were broken after repeated dilatation of the pupil with mydriatics and the signs of inflammation also ceasedto show. A similar phenomenon has also been observed previously during vector control (DADZIE et al., 1986). The regression of sclerosing keratitis was of only borderline statistical significance and exclusively concerned changes from advanced sclerosing keratitis to early sclerosing keratitis and to negative. Such major changes have not been reported previously, though some improvement in sclerosing keratitis after ivermectin treatment was seen in one study (TAYLOR et
al., 1989). Three casesof advanced sclerosing keratitis who becamenegative were, in fact, scored progressively until becoming negative, indicating progressive
improvement. Though it may be considered unimaginable that advanced sclerosing keratitis may regress, it is conceivable that in such a holo-endemic area with extremely
high ocular microfilarial
loads
which induce rapid development of eye lesions, a case of sclerosing keratitis, presenting with cornea1 haze as a result of cornea1 oedema, will be graded according to
271 the extent of the haze in the cornea. The cornea1 haze could gradually regress, following the elimination of the causative agent (microfilariae) from the cornea, as a result of the ivermectin treatment. This may account for some of the regressions observed. Photographic documentation, using sensitive equipment, might have provided the necessary confirmation of this observation. Lesions of the posterior segment of the eye remained stable throughout the 2-year study period and there was also no statistically significant change in the visual acuities of the population treated twice. However, 3 new cases of blindness occurred in the course of ivermectin treatment. These casesalready had eye lesions which had attained such a level of damage that the final outcome of blindness could no longer be prevented by ivermectin treatment. On the whole, the results of this study compared favourably with those obtained after 7 years of vector control (DADZIE et al., 1986), and showed the significant immediate benefit from community-based ivermectin treatment. It was not possible to demonstrate any appreciable regression of lesions of the anterior segment of the eye until 7 years after the start of vector control, when it was also observed that lesions of the posterior segment of the eye continued to progress (DADZIE et al., 1986). The significant regression of anterior segment lesions and the absenceof deterioration of eye lesions in general in persons treated twice, including a few individuals who tended intermittently to have very high ocular microfilarial loads, suggest that annual large-scale ivermectin treatment is adequate for the control of onchocerciasis. This contrasts with the opinion expressed in the previous report (DADZIE et al., 1990b) that 6-monthly treatment may be necessary to protect the populations in very highly endemic onchocerciasis areas. The logistic implication of this finding in the control of onchocherciasis cannot be overstated. Acknowledgement We are grateful to Dr Ebrahim M. Samba, Programme Director, whose continued encouragement and support made this report possible. References Anderson, J., Fuglsang, H., Hamilton, I’. J. S. & Marshall, T. F. de C. (1974). Studies on onchocerciasis in the United Cameroon Republic. II. Comparison of onchocerciasis in rain-forest and Sudan-savanna. Transactions of the Royal Society of Tropical Medicine and Hygiene, 68, 209-222.
Brown, K. R., Duke, B. 0. L., Ginger, C. D., Gutteridge, W. E., Le Berre, R., Overbosch, D., Remme, J., Samba, E. M., Van der Kaay, H. J. & Walsh, J. F. (1990). Proceedings of the symposium on onchocerciasis: conclusions and recommendations. Acta Leidensia, 59, 44+4X. Dadzie, K. Y., Remme, J., Rolland, A. & Thylefors, B. (1986). The effect of 7-8 years of vector control on the evolution of ocular onchocerciasis in West African s,t\anna. Tropical Medicine and Parasitology, 37, 26% L,“.
Dadzie, K. Y., Bird, A. C., Awadzi, A., Schulz-Key, H., Gilles, H. M. & A&, M. A. (1987). Ocular iindings in a double-blind study of ivermectin versus diethylcarbama-
zine versus placebo in the treatment of onchocerciasis. British Journal
of Ophthalmology, 71, 78-85.
Dadzie, K. Y., Awadzi, A., Bird, A. C. & Schulz-Key, H. (1990a). Ophthalmological results from a placebo controlled comparative g-dose ivermectin study in the treatment of onchocerciasis. Tropical Medicine and Parusizology, 40, 355-360.
Dadzie, K. Y., Remme, J., Alley, E. S. & De Sole, G. (1990b). Changes in ocular onchocerciasis four and twelve months after communitv-based treatment with ivermectin in a holoendemic onchocerciasisfocus. Transactions of the Royal So&~ Hygiene, 84, 103-108.
of Tropical Medicine and
De Sole, G., Remme, J., Awadzi, A., Accorsi, S., Alley, E. S., Ba, O., Dadzie, K. Y., Giese, J., Karam, M. & Keita, F. M. (1989a). Adverse reactions after large-scale treatment of onchocerciasis with ivermectin: combined results from eight community trials. Bulbin of the World Health Organization,
67, 707-719.
De Sole, G., Awadzi, A., Remme, J., Dadzie, K. Y., Ba, O., Giese, J., Karam, M., Keita, F. M. & Opoku, N. 0. (1989b). A community trial of ivermectin in the onchocerciasis focus of Asubende, Ghana. II. Adverse Fytions. Tropical Medicine and Parasiwlogy, 40, 375De Sole, G., Dadzie, K. Y., Giese, J. & Remme, J. (1990). Lack of adverse reactions from ivermectin treatment. Lancer, 335, 1106. Newland, H. S., Whyte, A. T., Greene, B. M,, D’Anna, S. A., Keyvan-Larijani, E., Aziz, M. A., Wilhams,P. N. & Taylor, H. R. (1988). Effect of single-doseivermectin on human Onchocercavolvulus infection with onchocercal ;;ly$volvement. BritishJournal of Ophthalmology, 72, Remme, J.,‘Baker, R. H. A., De Sole, G., Da*, K. Y., Walsh, J. F., Adams, M. A., Alley, E. S. & Avtssey, H. S. K. (1989a). A commumty trial of ivermectin in the onchocerciasis focus of Asubende, Ghana. I. Effect on the microfllarial reservoir and the transmission of Onchocercavolmdus. Tropical Medicine and Parasitology, 40, 367-374.
Remme, J., Dadzie, K. Y., Rolland, A. & Thylefors, B. (1989b). Ocular onchocerciasisand intensity of infection in the community. I. West African savanna. Tropical Medicine and Parasitology,
40, 340-347.
Remme, J., De Sole, G., Dadzie, K. Y., Alley, E. S., Baker, R. H. A., Habbema, J. D. F., Plaisier, A. P., van Oortmarssen?G. J. & Samba, E. M. (1990). Large-scale ivermectin hstribution and its epidemiological consequences. Acta Leidensia, 59, 177-190. Taylor, H. R., Murphy, R. P., Newland, H. S., White, A. T., D’Axma, S. A., Keyvan-Larijani, E., Aziz, M. A., Cupp, E. W. & Greene, B. M. (1986). Treatment of onchocerciasis. The ocular effects of ivermectin and ~~~y~arbamazine. Archives of Ophthalmoloe, 104,
--- -.-.
Taylor, H. R., Semba, R. D., Newland, H. S., KeyvanLariiani. E., White. A.. Dukulv. Z. & Greene. B. M. (1989). ~Ivermectin .treatment of patients with severe ocular onchocerciasis. American Journal of Tropical Medicine and Hygiene, 40, 494-500.
Thylefors, B. & Brinkmann, U. K. (1977). The microfilarial load in the anterior segment of the eye. A parameter of the intensity of infection of onchocerciasis.Bulletin of the World Health Organizati?
55, 731-737.
Vingtain, P., Pichard, E., Gmoux, J., Coulibaly, S. M., Bissan, Y., Ranque, P. & Thillaye, B. (1988). Ivermectine et onchocercose humaine: a propos d’une etude portant sur 234 onchocerquiens en republique du Mali. Bulletin de la Soctitb de Pathologie Exotique, 81,26&270. Received II September 1990; revised 12 November 1990; accepted for publication 14 November 1990
Changes in ocular onchocerciasis after ivermectin treatment in a holo-endemic K. Y. Dadzie, J. Remme and G. De Sole Ouagadougou, Burkina Faso
In a longitudinal study to determine the effect of annual community-based treatment of ocular onchocerciasis with ivermectin, the population living in the 3 most affected villages in the holo-endemic onchocerciasis focus of Asubende in Ghana were re-examined 16 and 24 months after initiating treatment . Ocular microfilarial loads had decreasedto very low levels in nearly all of the 334 examined persons who were treated twice. Only very few subjects had ocular loads of 32 microfilariae or more in the anterior chamber of the eye, but this was not associatedwith deterioration of ocular lesions. Important regression of both early and advanced lesions of the anterior segment of the eye was observed, which was highly statistically significant with respect to iridocyclitis. Lesions of the posterior segment of the eye remained stable. Though no systematic change in the visual acuity of the population was observed, 3 new casesof blindness occurred in persons who already had eye lesions at such. an advanced stage that ivermectin treatment could no longer affect the outcome, The results suggest that annual ivermectin treatment is adequate to control onchocercal ocular diseaseeven in populations with very high endemicity levels. Introduction
Ivermectin has been shown to be a safe microfilaritide suitable for large scale treatment of onchocerciasis (DE SOLEet d., 1989a, 1990; REMME ef al., 1990). However its full potential as a tool for the long-term control of onchocercal morbidity and the prevention of blindness is not yet known (BROWNet al.. 1990). In order to determine the effect of annual community-based ivermectin treatment on the eye, longitudinal studies were started in a population living in a holo-endemic onchocerciasis focus in Ghana. The first results, 4 and 12 months after the first treatment round, have been reported elsewhere (DADZIE et al., 1990b). In this paper, changes in ocular onchocerciasis 16 and 24 months after initiating annual community-based ivermectin treatment are reported.
The study population was drawn from the 3 most endemic communities in the isolated savanna onchocerciasis focus in the Asubende area in Ghana. where a community trial of ivermectin was started in 1987 (DE SOLEet al., 1989b; REMMEet al., 1989a). In these communities the prevalence of microfilariae in skin snips ranged from 85.6% to 87*3%, and the community microfilarial load from 58 to 73 microfilariae per skin snip. Of the total census population of 864 persons, 586 aged 5 years and over were examined
267
two rounds of community-based onchocerciasis focus
Onchocerciasis Control Programme in West Africa,
Abstract
Patients and Methods Study population
85, 267-271
B.P.
549,
ophthahnologically before the first mass treatment (DADZIE et al., 199Ob),and 581 and 586 persons were examined at the 16 months and 24 months surveys respectively. Drug administration
In October 1987 and October 1988 ivermectin tablets were given as a single oral dose of approximately 150 @kg body weight to all those in the study area who were eligible, under the direct supervision of a medical officer or a nurse. The exclusion criteria used during the second treatment were the same as durin the first treatment round, extent for women breast! eedina infants who were now excluded when breastfeeding &r infant of up to the age of 3 months instead of 1 month. The manufacturer of the drug had effected this change. Ophtkalnwlogical
examination
Detailed ophthalmological examination of all subjects aged 5 years and above was carried out before treatment and the examination was repeated 4,2 1, 16 and 24 months thereafter. The test of visual acuity was conducted using Sjiigren’s illiterate hand testtype in broad daylight. Slit lamp examination of the anterior segment of the eye, as well as fundus examination with the direct and indirect onhthalmoscope after pupil dilatation, was carried out in the onhthalmic mobile clinic followinn the onchocerciasis control project (OCP) protocol asieported previously (DADZIE et al., 1990a).All ophthalmological examinations were carried out by the same examiner and no reference was made to the record forms of previous examinations in the course of the survey. Classification of ocular lesions.
The 4 typical onchocercal eye lesions, viz. sclerosing keratitis, iridocyclitis, optic atrophy and choroido-retinitis. were classified into 3 erades: no lesion (absence of signs of any lesion), &rly lesion, and advanced lesion. A detailed description of the classification criteria has been given in the previous report (DADZIE et al., 1990a). Statistical analysis
Data processing was done using routine OCP computer proarams and SPSS/PCfor the final analvsis. 437 paher& had had a complete ophthalmological examination during the pre-treatment survey and those at 16 and 24 months after the start of the treatment programme. Of those patients, 334 had been treated twice with ivermectin with a yearly interval, 51 patients had been treated during the first treatment round only, 38 patients during the second round only, and 14 patients had been excluded from both treatments becauseof the exclusion criteria given
268 above. Only data from the right eye of the 14 non-treated patients and 334 patients treated twice were considered for the analysis. The trends in mean microfilarial loads during the full 2-year follow-up period could be shown only for the 289 patients who were treated twice and examined at all 5 surveys. Geometric mean microfilarial loads were calculated using the log (x+1) transformation. Differences in microfilarial loads were tested for statistical significance using the Wilcoxon signed rank sum test for comparisons between surveys. The 4 typical onchocercal eye lesions were allocated a score of 0 for no lesion, 1 for early lesion and 2 for advanced lesion. The significance of changes in ocular lesions within the treated and non-treated groups were tested using the Wilcoxon signed rank sum test on pre- and post-treatment scores for each lesion. For the comparison of the changes between the two groups the Mann-Whitney U test was applied to the difference between pre- and post-treatment scoresin caseswith a change in score. In 20 persons it was not possible to arrive at a classification for all lesions as the relevant signs had been obscured, usually by some other onchocercal pathology. In 18 subjects the view of the posterior segment was completely obstructed during at least one examination, and in 9 of them during all 3 examinations. For similar reasons it was not possible to obtain counts of microfilariae in the anterior chamber of the eye for 7 persons. These caseswith missing data were excluded from the analysis for the respective lesion or microfilarial load. 110 100
Results The Figure shows the trends in the mean number of microfilariae in the anterior chamber of the eye (MFAC) and in the mean number of dead microfilariae in the cornea (DMFC) of the 289 persons treated twice who were present at all 5 examinations. The mean microfilarial loads fell to low levels 4 months after each treatment round but showed a statistically significant increase 12 months thereafter (P
1st treatment
t
90 80 70 80 50 40 30 20 10 0 0
4
8
12
16
20
24
Number of months after first treatment Figure. Relative changes in ocular microfikarial loads after two rounds of ivermectin treatment; results for 289 patients; DMFC=mean dead microfihwiie in the cornea, MFAC=mean number of microfilariae in the anterior chamber.
number of
269 sion of anterior segment lesions. Table 2 shows the changes in distribution of onchocercal lesions of the anterior segment of the eye in patients treated twice. The classification of the presenceand severity of each lesion 16 and 24 months after the start of treatment is shown in relation to the pre-treatment classification of the sameeye in order to allow the assessmentof both random variation and systematic changes. At 16 months, 17 classified initially as having no sclerosing keratitis were reclassified as having an early lesion, but this was balanced by 19 cases with early sclerosing keratitis who were re-classified as having no lesion. Such results may be explained by random variation in the diagnosis of borderline lesions. The findings for advanced sclerosing keratitis are different. At 16 months only one new advanced lesion, which was initially classified as an early lesion, was reported, compared to 9 caseswith advanced lesions which had regressed, 6 to early lesions and 3 to no lesion. At 24 months, 14 of 19 cases with advanced sclerosing keratitis were re-classified as early or no lesion, and these changesin the treated group were of borderline statistical significance (P=O%).
The changesin the distribution of iridocyclitis were even more striking, and highly statistically significant. Of 36 caseswith early iridocyclitis pre-treatment, 32 were classified as negative after 24 months while not a single casemoved from negative to early lesion. Four caseswith advanced lesions were classified as negative for iridocyclitis at 24 months. With regard to choroido-retinitis, changes from no lesion to early lesioq, and from early lesion to advanced lesion, and vice versa, were similar and can be interpreted as random varmtion (Table 3). The same was true for optic atrophy. Table 3 shows that the pattern of ocular diseaseof the posterior segment of the eye remained stable during the lirst 2 years after the start of ivermectin treatment and that there was no statistically significant change during that period. The visual acuities of the 334 persons treated twice did not show any systematic change during the study period. However, 3 people in the study population had gone blind during the 24 months after the start of the community-based treatment, 1 after 12 months and the other 2 after 16 months. Two were men, aged 42 and 51 years respectively, who presented at the first examination with secondary glaucoma. Although
Table 1. Post-treatment distribution of microlihrial loads in the anterior chamber of the eye (MFAC) of 327 persons who were treated twice in relation to pre-treatment loads (percentage per pre-treatment group ia parentheses)
MFAC pre-treatment
No. of MFAC 16 months after star;;f trea3yTnt MFAC 24 months after start of treatment Persons 0 l40 l4lO32
0
144
l-
42
4-
31
lO-
78
32-
32
Total
327
$0)
$0,
(OVO)
(lf.0)
139 (96.5)
(2!8)
(017)
cove,
$0,
(9?2)
$4)
$0,
(2:)
$0)
(St;)
(9!5)
(2!4)
(2:)
(i-0,
(Io?O)
(,90,
(,90,
$0)
(090,
(7%)
(9.37)
(3!2)
$0)
(65.4)
(22.6) (O!O)
$3)
(6148s) (2?6)
(6!4)
(5!1)
(lf3)
(15!6)
(12:5)
(25!0)
(l:“-b,
(:!3)
$3)
(1~0)
(8;?7)
(S:!O) (1:*6) 301 (92.0) $4)
$5, (t8)
(9’4)
$5)
(f.2)
(k)
(3&) 256 (78.3)
$1)
$5) $5)
Table 2. Classification of onchocercal lesions of tbe anterior segment of the eye 16 and 24 months after the start of annual ivermectin mass treatment in relation to the pre-treatment classification
Type of lesion 16-monthfollow-up Sclerosing keratitis
Iridocyclitis 24-month follow-up
Sclerosing keratitis
Iridocyclitis
Post-treatment lesions Pre-treatment Non-treatedgroup Patients treated twice lesion None Early AdvancedNone Early Advanced Significanceof change None Early
10
2
-
;
i
; -
I 2
Advanced None Early Advanced
11
None Early Advanced None
11 10 1’:
1
-
7
;
-
-
2
Early Advanced
-
255 19 3 274 31 5
17 20 6
-1 10
-i -
; 14
253 22 3 274 32 4
19 16 11
-i 5
-i 2
13
-2
Within non-treated group f’ri*;T Within treated group Between the two groups P=O*34 Within non-treated group M&my Within treated group
Within non-treatedgroup P=l*O
Within treated group Between the two groups Within non-treated group Within treated group Between the two groups
P=OGl P=O*67 P=l-0 P
270 Table 3. Classilication of onchocercal ivermectin mass treatment in relation
Type of lesion Choroido-retinitis Optic atrophy
lesions of the posterior segment of the eye 24 months after the start of annual to the pre-treatment classification
Post-treatment lesions Pre-treatment Non-treatment group Patients treated twice None Early Advanced None Early Advanced Significance of change lesion None 10 269 6 Within non-treated group : I 6 7 -i Within treated group Early Advanced 6 13 Between the two groups 9 265 Within non-treated group None Early 1 ; ! 10 1; -4 Within treated group Advanced 1 3 16 Betweenthe two groups
they were treated symptomatically, the inevitable end of blindness could not be averted. The third was a 22 year old woman with advanced post-neuritic optic atrophy, for whom there was likewise no remedy available. The 2 men had been treated twice with ivermectin but the woman only once, as she had been excluded from the second treatment because of pregnancy. On the other hand, a 38 year old woman, who, because of corneal oedema resulting from bilateral secondary glaucoma could not count fingers at 3 m with either eye and who was by definition blind at the first examination, had responded to treatment with oral acetazolamideand topical steroids and mydriatics and was no longer classified as blind at the last surveys. She has also been treated twice with ivermectin .
Discussion
In order to determine the long-term impact of ivermectin treatment in the control of on&ocercal ocular disease. longitudinal oDhthahnoloeica1 followup examinations are needed td document carefully the changes in ocular disease occurring in communities included in mass treatment programmes (BROWN et al., 1990). Since the risk of developing onchocercal eye lesions is directly related to the intensity of infection, reflected by the microfilarial density in the eye and- in the skin (ANDERSONet al.; 1974; THYLEFORS& BRINKMANN. 1977: REMME et al.. 1989b), it is particularly imp&ant to do such studied in communities with high endemicity levels. In this respect the present study is unique as it reports the ocular changes after annual community-based treatment with ivermectin in a population with exceptionally high loads of ocular microfilariae. Ivermectin treatment resulted in a major reduction in ocular microfilarial loads 4 months after each treatment round. However, during the subsequent 8 months there was a sit&icant increase in the number of microfilariae in jhe eye, a phenomenon not observed in clinical trials (TAYLOR er al.. 1986: DADZIE et al., 1987, 1990a; ~VEWLANDet al:, 19881 VINGTAIN et al., 1988) but which is consistent with the post-treatment increase in skin microfilarial loads in the samepatients (REMME et al., 1990). Nevertheless, in most patients the ocular microfilarial loads remained very low, particularly after the second treatment, which resulted in a further reduction in the mean microfilarial loads in the eye. An exception to the above trend was observed in a few patients who, in spite of 2 ivermectin treatments, still had intermittently high loads of microfilariae in the anterior chamber of the eye. In none of them was
~$~ P=O.30 P=l.O P=O.34
P=O.78
worsening of the ocular diseaserecorded during the study period, while in half of them an improvement was observed. However, the number of casesinvolved was small and the period of observation relatively short. A longer period of observation, therefore, is necessaryto confirm whether intermittent high loads of ocular microfilariae in onchocerciasis patients under ivermectin treatment are unlikely to induce the development of eye lesions or cause the deterioration of existing ones. The previous report showed that early lesions of the anterior segment of the eyes tended to regress after the first round of ivermectin mass treatment and that the regression was particularly significant 4 months after treatment (DADZIE et al.. 1990a1.The nresent study not only cdnfirmed this oGervati& but px&ded further evidence that the regression of lesions is not only limited to early lesions of the anterior segment of the eye but involves advanced lesions also. The control group in this study, consisting of 14 subjects who had never been treated becauseof the exclusion criteria, was very small and this prevents a meaningful comparison with the group treated twice. Nevertheless, no tendency to regression among the non-treated patients was observed. The regression of iridocyclitis in treated patients was highly significant and, after 2 treatments, early iridocyclitis was hardly seen. Furthermore, as was noted in the previous report also (DADZIE et al., 1990b), there were several patients with iridocyclitis at the advanced stage of development who had regressedto negative when the synechiae were broken after repeated dilatation of the pupil with mydriatics and the signs of inflammation also ceasedto show. A similar phenomenon has also been observed previously during vector control (DADZIE et al., 1986). The regression of sclerosing keratitis was of only borderline statistical significance and exclusively concerned changes from advanced sclerosing keratitis to early sclerosing keratitis and to negative. Such major changes have not been reported previously, though some improvement in sclerosing keratitis after ivermectin treatment was seen in one study (TAYLOR et
al., 1989). Three casesof advanced sclerosing keratitis who becamenegative were, in fact, scored progressively until becoming negative, indicating progressive
improvement. Though it may be considered unimaginable that advanced sclerosing keratitis may regress, it is conceivable that in such a holo-endemic area with extremely
high ocular microfilarial
loads
which induce rapid development of eye lesions, a case of sclerosing keratitis, presenting with cornea1 haze as a result of cornea1 oedema, will be graded according to
271 the extent of the haze in the cornea. The cornea1 haze could gradually regress, following the elimination of the causative agent (microfilariae) from the cornea, as a result of the ivermectin treatment. This may account for some of the regressions observed. Photographic documentation, using sensitive equipment, might have provided the necessary confirmation of this observation. Lesions of the posterior segment of the eye remained stable throughout the 2-year study period and there was also no statistically significant change in the visual acuities of the population treated twice. However, 3 new cases of blindness occurred in the course of ivermectin treatment. These casesalready had eye lesions which had attained such a level of damage that the final outcome of blindness could no longer be prevented by ivermectin treatment. On the whole, the results of this study compared favourably with those obtained after 7 years of vector control (DADZIE et al., 1986), and showed the significant immediate benefit from community-based ivermectin treatment. It was not possible to demonstrate any appreciable regression of lesions of the anterior segment of the eye until 7 years after the start of vector control, when it was also observed that lesions of the posterior segment of the eye continued to progress (DADZIE et al., 1986). The significant regression of anterior segment lesions and the absenceof deterioration of eye lesions in general in persons treated twice, including a few individuals who tended intermittently to have very high ocular microfilarial loads, suggest that annual large-scale ivermectin treatment is adequate for the control of onchocerciasis. This contrasts with the opinion expressed in the previous report (DADZIE et al., 1990b) that 6-monthly treatment may be necessary to protect the populations in very highly endemic onchocerciasis areas. The logistic implication of this finding in the control of onchocherciasis cannot be overstated. Acknowledgement We are grateful to Dr Ebrahim M. Samba, Programme Director, whose continued encouragement and support made this report possible. References Anderson, J., Fuglsang, H., Hamilton, I’. J. S. & Marshall, T. F. de C. (1974). Studies on onchocerciasis in the United Cameroon Republic. II. Comparison of onchocerciasis in rain-forest and Sudan-savanna. Transactions of the Royal Society of Tropical Medicine and Hygiene, 68, 209-222.
Brown, K. R., Duke, B. 0. L., Ginger, C. D., Gutteridge, W. E., Le Berre, R., Overbosch, D., Remme, J., Samba, E. M., Van der Kaay, H. J. & Walsh, J. F. (1990). Proceedings of the symposium on onchocerciasis: conclusions and recommendations. Acta Leidensia, 59, 44+4X. Dadzie, K. Y., Remme, J., Rolland, A. & Thylefors, B. (1986). The effect of 7-8 years of vector control on the evolution of ocular onchocerciasis in West African s,t\anna. Tropical Medicine and Parasitology, 37, 26% L,“.
Dadzie, K. Y., Bird, A. C., Awadzi, A., Schulz-Key, H., Gilles, H. M. & A&, M. A. (1987). Ocular iindings in a double-blind study of ivermectin versus diethylcarbama-
zine versus placebo in the treatment of onchocerciasis. British Journal
of Ophthalmology, 71, 78-85.
Dadzie, K. Y., Awadzi, A., Bird, A. C. & Schulz-Key, H. (1990a). Ophthalmological results from a placebo controlled comparative g-dose ivermectin study in the treatment of onchocerciasis. Tropical Medicine and Parusizology, 40, 355-360.
Dadzie, K. Y., Remme, J., Alley, E. S. & De Sole, G. (1990b). Changes in ocular onchocerciasis four and twelve months after communitv-based treatment with ivermectin in a holoendemic onchocerciasisfocus. Transactions of the Royal So&~ Hygiene, 84, 103-108.
of Tropical Medicine and
De Sole, G., Remme, J., Awadzi, A., Accorsi, S., Alley, E. S., Ba, O., Dadzie, K. Y., Giese, J., Karam, M. & Keita, F. M. (1989a). Adverse reactions after large-scale treatment of onchocerciasis with ivermectin: combined results from eight community trials. Bulbin of the World Health Organization,
67, 707-719.
De Sole, G., Awadzi, A., Remme, J., Dadzie, K. Y., Ba, O., Giese, J., Karam, M., Keita, F. M. & Opoku, N. 0. (1989b). A community trial of ivermectin in the onchocerciasis focus of Asubende, Ghana. II. Adverse Fytions. Tropical Medicine and Parasiwlogy, 40, 375De Sole, G., Dadzie, K. Y., Giese, J. & Remme, J. (1990). Lack of adverse reactions from ivermectin treatment. Lancer, 335, 1106. Newland, H. S., Whyte, A. T., Greene, B. M,, D’Anna, S. A., Keyvan-Larijani, E., Aziz, M. A., Wilhams,P. N. & Taylor, H. R. (1988). Effect of single-doseivermectin on human Onchocercavolvulus infection with onchocercal ;;ly$volvement. BritishJournal of Ophthalmology, 72, Remme, J.,‘Baker, R. H. A., De Sole, G., Da*, K. Y., Walsh, J. F., Adams, M. A., Alley, E. S. & Avtssey, H. S. K. (1989a). A commumty trial of ivermectin in the onchocerciasis focus of Asubende, Ghana. I. Effect on the microfllarial reservoir and the transmission of Onchocercavolmdus. Tropical Medicine and Parasitology, 40, 367-374.
Remme, J., Dadzie, K. Y., Rolland, A. & Thylefors, B. (1989b). Ocular onchocerciasisand intensity of infection in the community. I. West African savanna. Tropical Medicine and Parasitology,
40, 340-347.
Remme, J., De Sole, G., Dadzie, K. Y., Alley, E. S., Baker, R. H. A., Habbema, J. D. F., Plaisier, A. P., van Oortmarssen?G. J. & Samba, E. M. (1990). Large-scale ivermectin hstribution and its epidemiological consequences. Acta Leidensia, 59, 177-190. Taylor, H. R., Murphy, R. P., Newland, H. S., White, A. T., D’Axma, S. A., Keyvan-Larijani, E., Aziz, M. A., Cupp, E. W. & Greene, B. M. (1986). Treatment of onchocerciasis. The ocular effects of ivermectin and ~~~y~arbamazine. Archives of Ophthalmoloe, 104,
--- -.-.
Taylor, H. R., Semba, R. D., Newland, H. S., KeyvanLariiani. E., White. A.. Dukulv. Z. & Greene. B. M. (1989). ~Ivermectin .treatment of patients with severe ocular onchocerciasis. American Journal of Tropical Medicine and Hygiene, 40, 494-500.
Thylefors, B. & Brinkmann, U. K. (1977). The microfilarial load in the anterior segment of the eye. A parameter of the intensity of infection of onchocerciasis.Bulletin of the World Health Organizati?
55, 731-737.
Vingtain, P., Pichard, E., Gmoux, J., Coulibaly, S. M., Bissan, Y., Ranque, P. & Thillaye, B. (1988). Ivermectine et onchocercose humaine: a propos d’une etude portant sur 234 onchocerquiens en republique du Mali. Bulletin de la Soctitb de Pathologie Exotique, 81,26&270. Received II September 1990; revised 12 November 1990; accepted for publication 14 November 1990