SEARCH

SEARCH BY CITATION

Keywords:

  • neurolathyrism;
  • Ethiopia;
  • epidemic

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

OBJECTIVES To describe the neurolathyrism epidemic in Ethiopia and to identify associated household factors.

METHODS We interviewed 589 randomly selected heads of household in Debre Sina district of Ethiopia, the area afflicted by the recent neurolathyrism epidemic. Disease information was obtained for 2987 family members.

RESULTS Neurolathyrism patients were detected in 56 (9.5%) households (prevalence rate 2.38%). The mean number of affected family members per household was 1.27 (SD 0.65, range 1–3). Most (77.5%) patients developed the disability during the epidemic (1995–1999). The median age at onset of paralysis was 11 years with a range of 41 (range 3–44). Younger people were more affected during the epidemic than during the non-epidemic period (P=0.01). The presence of a neurolathyrism patient in the family was associated with illiteracy [adjusted OR (95% CI)=2.23 (1.07–5.10)] of the head of household, with owning a grass pea farm [adjusted OR (95% CI)=2.01 (1.04–3.88)] and with the exclusive cooking of grass pea foods using handmade traditional clay pots [adjusted OR (95% CI=2.06 (1.08–3.90)].

CONCLUSION Males aged 10–14 years were most affected by neurolathyrism. Increased household risk was associated with illiteracy of the head of the household and exclusive cooking of grass pea foods with handmade traditional clay pots.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Neurolathyrism is an ancient disease known since the time of Hippocrates who wrote that continuous consumption of peas could lead to `impotence' in the legs (Cohn & Kislev 1987). An ancient Hindu writing also stated grass pea as a cause of lameness (Weber 1998). In 1671, the Duke of Württemberg in Germany issued a decree that prohibited the baking and consumption of bread containing grass pea (Cohn & Streifler 1978).

Neurolathyrism is an upper motorneurone disease caused by over-consumption of the drought-tolerant grass pea, Lathyrus sativus, that contains the glutamate analogue neurotoxin β-N-oxalyl-α,β-diaminopropionic acid (β-ODAP), also known as β-oxalylaminoalanine (BOAA) (Spencer et al. 1986). The toxin is believed to mediate its neuronal damage through its excitation of the AMPA subtype of the ionotropic glutamate receptor (Spencer 1999). However, the toxin also has largely non-specific binding characters with synaptic membranes (Jain et al. 1998) and other neurophysiological activities are exhibited by β-ODAP and by its de-oxalylation product α,β-diaminopropionic acid (Kusama et al. 2000).

Females are relatively rarely affected; dietary habits (Haque et al. 1996) and hormonal factors (Dwivedi & Prasad 1964; Haimanot et al. 1990) have been suggested as aetiological factors. Blood group O may increase the susceptibility to neurolathyrism (Getahun et al. 1999b). The incidence varies from village to village even during epidemics (Haimanot et al. 1993). The disease is characterized by neuronal death, it is incurable and disability is permanent.

Neurolathyrism often occurs as epidemics during times of food shortages after drought and flooding. Several epidemics appeared in many parts of the world including Europe, China and the Indian subcontinent (Weber 1998). During the Second World War, in a German forced labour camp on Ukrainian territory, the Rumanian Jewish inmates were given 400 g of grass pea boiled in salt water with an additional 200 g bread made of barley (80%) and chopped straw (20%). Of the 1350–1400 inmates, 800 developed paralysis within 3 months (Kessler 1947). The most recent neurolathyrism epidemic was reported from northeastern Ethiopia in 1999 where drought led to famine and excessive consumption of the drought-tolerant grass pea (Getahun et al. 1999a). This epidemic, which was detected in 1997, afflicted a number of grass pea growing districts in this region and continued through 1999 because of persistent drought (Health Bureau 1999). The objective of this study, conducted in March 2000, was to describe the epidemic and identify associated household factors.

Subjects and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The study was conducted in Debre Sina district of South Wello Zone of the Amhara region, Ethiopia, which is one of the districts that has been afflicted by the recent neurolathyrism epidemic. The district is divided into 31 kebeles (smallest administrative unit) and had an estimated population of 150 000 in 2000. Four rural kebeles that were afflicted by the neurolathyrism epidemic were randomly selected to be included into the study. Collectively there were 5890 households in these four kebeles. This number was used to estimate a sample of 10% from each. The first household near a clinic in the kebele or a main gravel road was used as a starting point to enrol study subjects in a systematic simple random sampling until the required quota of households was achieved. Using this method, 2987 inhabitants of 589 households were included into the study. This sample size was larger than necessary, considering an estimated prevalence of neurolathyrism of 2%, a precision of 0.5% and a confidence level of 95% (2952 subjects).

Trained high school graduates interviewed heads of household using a closed- and open-ended questionnaire designed to collect information on sociodemographic variables and proxy variables for wealth, such as land, cattle ownership and housing conditions. The questionnaire was pretested outside the study area. Absent respondents and respondents with incomplete data were revisited at home to obtain full information. Trained nurses and the principal investigator supervised the data collection in the field and the home revisits. There were no refusals to participate in the study. Ethical clearance was obtained from the Ethiopian Science and Technology Commission and Ghent University, Belgium.

The case definition for neurolathyrism was symmetrical spastic leg weakness, subacute or insidious onset, with no sensory deficit and history of grass pea consumption prior to and at the onset of paralysis. The principal investigator, a physician with extensive field experience of the disease, examined all neurolathyrism patients. The case definition has a sensitivity of 95% and a specificity of 90% in detecting neurolathyrism patients (Haimanot et al. 1990).

The degree of disability of patients was categorized into four groups as follows:

Stage I: Spastic gait with no need to use a walking stick and no Babinski sign (Acton's no stick stage);

Stage II: Spastic gait with the need of one walking stick. Ankle clonus and Babinski sign present (one-stick stage);

Stage III: Spastic cross adductor gait with the need of two walking sticks. Ankle clonus and Babinski sign present (two-stick stage);

Stage IV: Bedridden with loss of leg use and contracture. Arms are strong and pyramidal signs are present (crawler stage).

Descriptive statistics were used for socio-demographic and other household related variables. Presence or absence of a neurolathyrism patient was used as a household measure of the risk of neurolathyrism and as a dependent variable both in univariate and multivariate analysis. Pearson's chi-square test was used to compare proportions. Fisher's exact test was used when Pearson's chi-square test was not applicable. Increasing age of household head and family size favour presence of a neurolathyrism patient in the household; hence both were entered into a logistic regression model to obtain the adjusted odds ratio, associated 95% confidence interval and the P-values (Wald chi-square). Those neurolathyrism patients who developed the paralysis between 1995 and 1999 were grouped as epidemic cases. During analysis a new variable (severity) was created. The first and second stages of the disease were coded as less severe and the third and fourth stages as more severe. Means were compared by independent sample t-test; significance of the P-value was set at the two-sided level of 0.05. SPSS statistical package version 9 was used for data analysis.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Among 2987 individuals in 589 households, 71 [prevalence rate (95% CI)=2.38% (1.83–2.93%)] neurolathyrism patients were detected in 56 (9.5%) of the households. The mean number of affected family members per household was 1.27 with a range of three (maximum four and minimum one). The majority [46/56 (82.1%)] of the affected families had a single affected family member. This was followed by two patients in 10.7%, three patients in 5.4% and four patients in 1.8% of the affected families. The median age of the patients was 13 years with a range of 62 (maximum 68 and minimum 6). Fifty-four (76%) patients were younger than 20, and 70.4% (50/71) were males with a male-to-female ratio of 2.3 : 1.

The earliest year of onset of paralysis reported was 1972 and the latest 1999. Figure 1 shows the distribution of cases by year of onset and gender. Most patients [55 (42 males and 13 females), male-to-female ratio 3.2 : 1] developed the disability during the epidemic (1995–1999).

image

Figure 1.  The distribution of neurolathyrism cases by gender and year of onset of paralysis, Debresina, Ethiopia, March 2000.

Download figure to PowerPoint

The second stage of the disease was the most common and found in 33 (46.5%) of patients, followed by first (36.6%), fourth (9.9%) and third stage (7%). There was no association between the severity of the disease and gender (P=0.29). Likewise, there was no difference in severity of disease between epidemic and non-epidemic cases.

The median age at onset of paralysis was 11 years with a range of 41 (maximum 44 and minimum 3). There was no association between age at onset of paralysis and gender (t-test=0.23; P=0.82). The association between age at onset of paralysis and severity of disease was borderline non-significant (P=0.05). The median age at onset of paralysis was 10 years for epidemic cases and 19 for non-epidemic cases; this difference was statistically significant (t-test=2.84; P=0.01). In this preliminary survey, the age and gender of all family members could not be obtained. As an alternative, data from the most recent census were used to calculate ratios. Table 1 shows the 5-year age group distribution at onset of paralysis between total, epidemic and non-epidemic cases compared with the proportion of the rural zonal population of South Wello as a reference population. South Wello administrative zone encompasses the study district and 16 others. The age group 10–14 years was most susceptible among the epidemic cases, but there was no clear age-related susceptibility among non-epidemic cases (Table 1), nor did the severity of disease (P=0.82) differ between epidemic and non-epidemic cases.

Table 1.   Distribution of age groups at onset of paralysis of all neurolathyrism patients (n=71) and cases during the epidemic (n=55) compared with a reference population, Debre Sina, Ethiopia, March 2000 Thumbnail image of

The majority (85.9%) of the responding heads of household were male; their mean age was 44.1 years (SD 12.8) with a range of 65 years (maximum 83 and minimum 18). Almost all (99.7%) were farmers. The mean family size was 5 (SD 2, range 1–13). Most (64.5%) were illiterate and married (87.3%). More Muslim heads of household were illiterates (P < 0.001).

Table 232 shows the distribution and unadjusted OR of selected variables among neurolathyrism-affected (n=56) and non-affected (n=533) families. Presence of a neurolathyrism patient in the family was associated with illiteracy [unadjusted OR (95% CI)=2.74 (1.35–5.55)] and Islam religion of the head of the household [unadjusted OR (95% CI)=2.21 (1.14–4.29)], with owning a grass pea farm and with exclusive cooking of grass pea foods using traditional clay pots.

Table 2.   Distribution and unadjusted OR of selected variables by neurolathyrism affected (n=56) and non-affected (n=533) households, Debre Sina, Ethiopia, March 2000 Thumbnail image of
Table 3. Table 2 (Continued)Thumbnail image of

After adjustment for the age of the head of household and the family size of the household in a logistic regression model, illiteracy [adjusted OR (95% CI)=2.23 (1.07–5.10)] and Islam religion [adjusted OR (95% CI)=2.85 (1.27–6.38)] of the head of household, currently owning a grass pea farm [adjusted OR (95% CI)=2.01 (1.04–3.88)] and exclusive cooking of grass pea foods using handmade traditional clay cooking pots [adjusted OR (95% CI)=2.06 (1.08–3.90)] remained significant (Table 3). In the logistic regression model, the age of the head of household and the family size were non-significant.

Table 3.   Distribution and adjusted OR (for the age of the head of the household and the family size) of variables by neurolathyrism affected (n=56) and non-affected (n=533) households, Debre Sina, Ethiopia, March 2000 Thumbnail image of

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This is the third epidemic reported in the same region following reports by Rizzotti (1952) and Gebreab et al. (1978). It has followed a famine caused by drought (Getahun et al. 1999), which is similar to the previous reports. The popularity of grass peas, both as cash and subsistence crop, is increasing in the drought-prone areas of the country, and it is over-consumed by the community at times of food shortages (Haimanot et al. 1993; Getahun & Haimanot 1998).

The prevalence of neurolathyrism in our study (2.38%) was similar to the rate of 2.44% during the previous epidemic in 1976 (Gebreab et al. 1978), but it is higher than prevalence rates from north-west Ethiopia (Haimanot et al. 1990, 1993). As these previous epidemiological studies were conducted more than 10 years after the epidemics, it is likely that some cases were lost. Cohn (1994) found that in countries with excellent medical care neurolathyrism does not affect the longevity of the victims, but this may be different in countries with very poor health care, as in Ethiopia. After the recent neurolathyrism epidemic there, victims of the disease have migrated to the bigger cities to make a living as beggars (Getahun 2000). By contrast our study was conducted in the aftermath of a recent epidemic. Neurolathyrism is manifesting in Ethiopia with an estimated annual incidence rate of 1.7 per 10 000 (Haimanot et al. 1990).

The overall sex and age distribution of patients in our study was similar to most of the previous studies in Ethiopia (Gebreab et al. 1978; Haimanot et al. 1990; Getahun & Haimanot 1998) and elsewhere (Dwivedi & Prasad 1964; Haque et al. 1996). In our study the male preponderance was demonstrated more among the epidemic (1995–1999) than the non-epidemic cases. Previous studies showed that females have earlier onset of paralysis and less severe symptoms (Haimanot et al. 1993; Getahun & Haimanot 1998). However, in our study neither age at onset of paralysis nor severity of disease were associated with gender.

Most previous studies in Ethiopia (Haimanot et al. 1993) or elsewhere (Dwivedi & Prasad 1964; Haque et al. 1996) used 10-year age groups, and consistently found that those aged 10–20 years were most often affected. Gebreab et al. (1978) reported the age group of 6–10 as that most affected during the previous neurolathyrism epidemic of Ethiopia in 1976. We found that when standardizing the 5-year age group distribution at onset of paralysis with the proportion of the rural population of South Wello Administrative zone, children aged 10–14 were affected most among the epidemic cases. This age group coincides with the period of pubertal growth spurt and the corresponding increased requirement for calorie intake may have resulted in greater consumption of grass pea. We found that 71.8% of all cases and 83.6% of epidemic cases developed paralysis between 5 and 14 years, which clearly indicates that the disease affects the youngest section of the community. Non-epidemic cases neither show a sharp peak of age at onset of paralysis nor male preponderance of cases. But a firm conclusion cannot be drawn because of the smaller number of non-epidemic cases.

The findings of our study in terms of the stages of the disease are different from previous studies. The second stage of the disease was the commonest disability and the most severe form of the disease (the fourth stage) was proportionally more frequent than in previous reports (Gebreab et al. 1978; Haimanot et al. 1990). Likewise, the number of patients per family was higher than in previous studies in Ethiopia (Haimanot et al. 1990) and Bangladesh (Haque et al. 1996).

Another striking finding of our study was that families whose head was illiterate carried an increased household risk of paralysis. Education drives both individual and community development, and illiterates are likely to have low socio-economic status even in such remote rural areas. Neurolathyrism is a disease of the poor (Haimanot et al. 1993; Haque et al. 1996). Although home detoxifying methods such as parboiling and aqueous leaching of the toxin of grass pea have long been known (Jha 1987), the rural communities were not familiar with them (Getahun & Haimanot 1998). Literate people probably have better access to information about home detoxification methods, but for the explicit understanding of this issue future studies are needed.

The Islam religion as a risk factor is difficult to interpret in this study. In another district of the region cases peaked among Coptic Orthodox Christians, coinciding with the end of a 2-month fasting period during which animal products were not consumed (Getahun et al. 1999a). Residual confounding by illiteracy and socio-economic status may be a possible explanation in our study. Both were important predisposing factors and also strongly associated with religion. Socioeconomic status in particular was only indirectly controlled for in the analysis.

Exclusive cooking of grass pea foods with handmade traditional clay pots, which can leach iron into food (Adish et al. 1999), was associated with household risk of paralysis. Studies with human subjects and experimental animals suggest that disturbance in the regulation of iron metabolism, iron induced oxidative stress and free radical formation are major pathogenic factors in neurodegenerative disorders (Jellinger 1999). The symmetrical axonal degeneration of both crossed and uncrossed pyramidal tracts in the spinal cord is the principal pathological change observed in human neurolathyrism (Streifler et al. 1977). Further biochemical studies are needed addressing the leaching of iron or other micronutrients that could have a role in the pathogenesis of neurolathyrism from traditional clay pots.

Finally, it should be stressed that excessive consumption of grass pea is the single most predisposing factor of neurolathyrism. Moreover, it was found that causation of the disease is poorly understood both by the population and by health workers working in neurolathyrism-prone areas (Getahun et al. 2002). Therefore, appropriate health information, communication and training has to be provided.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors are grateful to Dr Kunuz Abdella and Dawit Beyene for their help during the data collection. We thank Dr Dirk De Bacquer, statistician at the Public Health Department of the Ghent University, Belgium for his constructive comments. The study was partially funded by the Belgian Technical Co-operation (BTC). Last but not least our gratitude goes to the supervisors, enumerators and the rural communities who participated in the study.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Adish AA, Esrey SA, Gyorkos TW, Jean-Baptiste J & Rojhani A (1999) Effect of consumption of food cooked in iron pots on iron status and growth of young children: a randomised trial. Lancet 353, 712716.
  • 2
    Cohn DF (1994) Does neurolathyrism affect longevity? In: The Grass Pea and Lathyrism (eds BM Abegaz, RT Haimanot, VS Palmer & Spencer PS). Proceedings of the Second International Lathyrus/Lathyrism Conference in Ethiopia. Third World Medical Research Foundation, New York, pp. 33–35.
  • 3
    Cohn DF & Kislev ME (1987) Human neurolathyrism. Adler Museum Bulletin 13, 58.
  • 4
    Cohn DF & Streifler M (1978) Neurolathyrism. Historical notes. Koroth 7, 147152.
  • 5
    Dwivedi MP & Prasad VG (1964) An epidemiological study of lathyrism in the district of Rewa, Madhya Pradesh. Indian Journal of Medical Research 52, 81114.
  • 6
    Gebreab T, WoldeGabriel Z, Maffi M, Ahmed Z, Ayele T & Fanta H (1978) Neurolathyrism. A review and report of an epidemic. Ethiopian Medical Journal 16, 111.
  • 7
    Getahun H (2000) Lathyrism in Ethiopia: an un-addressed problem. Lathyrus Lathyrism Newsletter 1, 78.
  • 8
    Getahun H & Haimanot RT (1998) Psychosocial assessment of Lathyrism patients in rural Estie district of south Gonder, northern Ethiopia. Ethiopian Medical Journal 36, 918.
  • 9
    Getahun H, Lambein F & Vanhoorne M (2002) Neurolathyrism in Ethiopia: assessment and comparison of knowledge and attitude of health workers and rural inhabitants. Social Science and Medicine 55 (in press).
  • 10
    Getahun H, Mekonnen A, TekleHaimanot R & Lambein F (1999a) Epidemic of neurolathyrism in Ethiopia. Lancet 354, 306307.
  • 11
    Getahun H, Serniclaes W & Lambein F (1999b) The role of ABO blood groups in the susceptibility of the toxic neurodegenerative disorder neurolathyrism in Ethiopia. X Annual Scientific Conference of the Ethiopian Public Health Association, October 1999. Addis Ababa, Ethiopia. Abstract 25, p. 26.
  • 12
    Haimanot RT, Kidane Y, Wuhib E et al. (1990) Lathyrism in rural northern Ethiopia: a highly prevalent neurotoxic disorder. International Journal of Epidemiology 19, 664672.
  • 13
    Haimanot RT, Kidane Y, Wuhib E et al. (1993) The epidemiology of Lathyrism in north and Central Ethiopia. Ethiopian Medical Journal 31, 1524.
  • 14
    Haque A, Hossain M, Wouters G & Lambein F (1996) Epidemiological study of Lathyrism in Northwestern districts of Bangladesh. Neuroepidemiology 15, 8391.
  • 15
    Health Bureau (1999) Lathyrism Epidemic, Field Report. Health programs department, Amhara National Regional State, Bahir Dar, Ethiopia.
  • 16
    Jain RK, Junaid MA & Rao SLN (1998) Receptor interaction of beta-N-oxalyl-L-alpha,beta- diaminopropionic acid, the Lathyrus sativus putative excitotoxin, with synaptic membranes. Neurochemical Research 23, 11911196.
  • 17
    Jellinger KA (1999) The role of iron in neurodegeneration: Prospect for pharmacotherapy of Parkinson's disease. Drugs and Aging 14, 115140.
  • 18
    Jha K (1987) Effect of the boiling and decanting method of Khesari (Lathyrus sativus) detoxification, on changes in selected nutrients. Archives of Lathinoam Nutrition 37, 101107.
  • 19
    Kessler A (1947) Lathyrismus. Psychiatrie und Neurologie 113, 345376.
  • 20
    Kusama T, Kusama-Eguchi K, Ikegami F et al. (2000) Effects of β-ODAP, the Lathyrus sativus neurotoxin, and related natural compounds on cloned glutamate receptors and transporters expressed in Xenopus oocytes. Research Comm in. Pharmacology and Toxicology 5, 3755.
  • 21
    Rizzotti G (1952) Lathyrismo in Ethiopia. Bulletino Della Societa Italiana Di Medicine and Igiene Tropicale 33, 493500.
  • 22
    Spencer PS (1999) Food toxins, AMPA receptors and motor neuron diseases. Drug Metabolism Reviews 31, 561587.DOI: 10.1081/DMR-100101936
  • 23
    Spencer PS, Roy DN, Ludolph A, Hugon J, Dwivedi MP & Shaumburg HH (1986) Lathyrism. Evidence for role of neuroexcitatory amino acid BOAA. Lancet ii, 10661067.
  • 24
    Streifler M, Cohn DF, Hirano A & Schuman E (1977) The central nervous system in a case of neurolathyrism. Neurology 27, 11761178.
  • 25
    Weber KT (1998) A wild vetchling's ominous tendrils gone to the loins. Cardiovascular Research 37, 551553.