15

Redda Tekle Haimanot, Yemane Kidane, Elizabeth WüTííb, Angelina Kassina,Yohannes Endeshaw, Tadesse Alemu, Peter S. Spencer. 1993. Ethiop Med J 31,

THE EPIDEMIOLOGY OF LATHYRISM IN NORTHAND CENTRAL ETHIOPIA

Redda Tekle Haimanot, MD, FRCP(C), PhD1, Yemane Kidane,MA2, Elizabeth Wuhib, BA2, Angelina Kassina, Dip. Nut2,

Yohannes Endeshaw, MD3, Tadesse Alemu, MD, MSc1,Peter S. Spencer, PhD, MRC Path4.

ABSTRACT: Lathyrism is a neurotoxic disorder caused by excessive, prolongedconsumption of the hardy, environmentally tolerant legume, the grass-pea, Lathy russativus, which contains the neurotoxic amino acid beta-N-oxalylamino-L-alanineacid (BOAA). The disease develops after heavy consumption of grass-pea for overtwo months. It is uniformly manifested by a predominantly motor spasticparaparesis with varying degrees of disability. A door-to-door epidemiologicalsurvey for the disease using trained lay health workers was carried out in the majorareas of northwest and central Ethiopia where L. sativus is grown. For securityreasons, some of the other endemic areas were not accessible for the survey. Thesurvey involved a population of 1,011,272. A total of 3,026 affected persons wereidentified. The disease was found to be widespread in the northwest and centralhighland areas of the country. The prevalence rates ranged from 1/10,000 to7.5/1,000. The highest prevalences were in North and South Gonder, and East andWest Gojam. The male.female ratio of cases was 2.6:1; the females exhibited amilder form of the disease. The cultivation ofL. sativus is increasing in Ethiopia,which makes the development of low-BOAA strains very important in order tocontrol the high incidence of lathyrism, a crippling disease which affects theproductive young members of the society.

1 Faculty of Medicine, P. O. Box 4147, Addis Abeba University, AddisAbeba, Ethiopia.

Ethiopian Nutrition Institute, Addis Abeba, Ethiopia.

3 Gonder College of Medical Sciences, Addis Abeba University, P.O. Box 196,Gonder, Ethiopia.

4 Centre for Research on Occupational and Environmental Toxicology, OregonHealth Sciences Universtiy, Portland, Oregon 97201, USA.

16

INTRODUCTION

Lathyrism is a neurotoxic disordercaused by excessive, prolonged con-sumption of grass-pea (Lathyrus sativu-s). Much of our knowledge about thispulse and the disabling disease it causescomes from studies in India (1-4). Theneurotoxic culprit, beta-N-oxalylamino-L-alanine acid (BOAA) has been ident-ified and characterized (5). Although inthe past the disease occurred in Europe,North Africa, the Middle East and Asia(6), it is at present endemic in India(7), Bangladesh (8) and Ethiopia (9). InEthiopia, it has been known for manyyears. Its leg-paralysing features havegained descriptive local names such as"sebre", "gayu" and "gwaya". Althoughknown to be associated with L. sativus,the paralysis has given rise to falselocal beliefs concerning the method bywhich the disease is acquired. Out-breaks of lathyrism have occurred inthe northern parts of the country associ-ated with famine (10, 11), and thedisorder remains endemic in severalparts of the country, with the occur-rence of new cases recently recorded inthe northwest (12).

The purpose of this study is to givea general epidemiological overview ofthe lathyrism problem in the endemicregions of Ethiopia with the aim ofproviding information essential for theprevention and possible eradication ofthe disabling disease. It was undertakenas part of a multidisciplinary approachto study the agricultural, medical andnutritional aspect of the disorder (13).Study areas (Figure 1): Grass-pea (L.

sativus) is grown in the Ethiopian high-lands at altitudes ranging from 1700 to2220 m and a maximum temperature of15 - 20 C. Its production is mainly inareas affected by severe natural disas-ters such as moisture stress, waterlogging and flooding. Grass-pea cultiva-tion is concentrated predominantly inthe northwest zone (58%), while thecentral and northeast zones produce16.3% and 12.8% respectively. Thenorthern and southeast regions producethe balance (14).The survey: Field investigations carriedout between 1988 and 1990 were con-centrated on the major grass-pea pro-ducing administrative regions of Northand South Gonder, West and EastGojam and West Shewa. For securityreasons, the survey was unable to coverWelo, Tigray, Eritrea, Harer, and partsof Shewa Administrative Regions.There were also inaccessible grass-peagrowing districts in North and SouthGonder.

In each administrative region, theresearch team contacted the regionalofficers of the Ministry of Agricultureas well as the local administrative auth-orities responsible for crop production.Up-to-date information on grass-peacultivation and production wasobtained. A twenty-per-cent randomsampling of the Peasant Associations ofgrass-pea cultivating districts was made.Later on in the survey, the randomsampling had to be reduced to 10%,due to security considerations and theneed to complete the survey before themajor military escalations of late 1991in the survey areas. The results of a

neuro-epidemiology survey from theButajira district of South Shewa, aregion which does not grow grass-pea(15), were included for control purpose.

Locally recruited lay health workers(6 males and 4 females) were trained inthe use of questionnaires and the recog-nition of motor disabilities in the legs,including the different stages ofparaparesis due to lathyrism. Pilotstudies were undertaken in every surveyarea whenever a new group ofenumerators was recruited.

Village elders and leaders of thedifferent Peasant Associations wereinformed about the objectives of thefieid investigation, and their consentwas obtained. The elders were inter-viewed individually and in groups onthe dietary habits of the community,with particular reference to grass-peaconsumption.

During their door-to door interviews,the enumerators collected census dataincluding age, sex, religion, maritalstatus, occupation and food habits ofhouseholds in each village. They alsorecorded information on the amount andperiod of grass-pea consumption inhouseholds, and the onset of walkingdifficulties. In the majority of cases, theage of individuals and the age of onsetof symptoms were determined usinglocal event calendars. Two physicianstrained to recognize and diagnose casesof lathyrism screened all subjects withwalking difficulties identified by theenumerators. The neurologist in theteam saw sampled groups of cases inorder to validate diagnosis. After aperiod of experience, the agreement

17

between the physicians and neurologistswas excellent.

Inclusion criteria included leg weak-ness and spasticity of subacute or insidi-ous onset and history of heavy grass-pea consumption prior to and at onsetof the disease. The following wereconsidered to be exclusion criteria:sensory deficit in extremities, markedasymmetrical leg weakness, continuingbladder and/or bowel dysfunction, andpresence of peripheral gangrene in thelimbs (12). Resulting disability wasclassified according to Acton (2) withsome modifications.

Stage 1: mild spastic gait with nouse of a stick; increased stiffness andexaggerated deep tendon reflexes (DTR)of die lower limbs; ankle clonus pres-ent, Babinski's sign present.

Stage 2: spastic gait, with use of onestick; lower limbs exhibit increasedstiffness; mild rigidity; exaggeratedDTR; ankle clonus present, Babinski'ssign present.

Stage 3: spastic gait, with use of twosticks; crossed adductor gait; lowerlimbs exhibit exaggerated DTR; ankleclonus present; Babinski's sign present.

Stage 4: crawling or bedridden state;loss of leg use, with contracture; armsstrong; and pyramidal signs present.

Concurrent with the medico-socialfield surveys, the team collectedsamples of grass-pea from the localmarket and some of the households. Atotal of 200 grass-pea seed and foodsamples collected from the surveyedareas were sent to the Addis AbebaUniverstiy's Department of Chemistryfor analysis of BOAA content using the

18

calorimetric ophthaldehyde (OPT)fluorescent dye method (16, 17).Statistical Methods: the two-tailed testwas used to test for statistical signifi-cance.

RESULTS

Table 1 shows the 1988 production andyield of grass-pea cultivation inEthiopia as depicted by the CentralStatistics Office (CSO) (18). East Go-jam, South Gonder, West Gojam, andSouth Welo are, in that descendingorder, the major grass-pea producingregions, followed by East Shewa, NorthWelo, North Gonder and Arsi. Beta-N-oxalylamino-L-alanine acid (BOAA)levels in the 200 samples collected fromthe survey areas ranged from 0.2 to0.96 mg/100gm of dry grass pea. HighBOAA content was registered fromgrass-pea samples collected from Gon-der and Gojam. The highest were fromthe sub-district of Adet (West Gojam).The grass-pea samples in Shewa are ingeneral less toxic. The prevalence oflathyrism was highest in the subdistrictsof Yilmana and Deinsa (West Gojam),Dembia (North Gonder) and Fogera(South Gonder).

In all the lathyrism endemic areas,agriculture, including cattle rearing,was the main occupation, and the chiefsource of livelihood, for over 90% ofthe affected persons. Socio-economical-ly the population was poor (i.e. belowthe average Ethiopian per capita GNPof 120 US dollars per year) and individ-uals had a hand-to-mouth existence.

The age and sex distribution (Fig. 2)

demonstrates that males were affectedmuch more than females (2.6: 1). Onsetof the disease took place below 10years of age in 26.7%, between 10 and20 years in 27.1%, between 21 and 40years in 29.9%, and above 40 years in16.3 per cent. Thus, 83.7% were belowthe age of 40 years at onset. It is note-worthy that, in females, 69.2% of caseshad their disease onset before the age of20 years as compared to 47.7% inmales (p<0. 001).

In over 80% of the cases, the diseasedeveloped in the months of May toAugust (the majority in July), after 3 to6 months of grass-pea consumption in65%, 2 to 3 months in 60% and onemonth or less in 15%. In Ethiopia,grass-pea is sown in October/Novemberand harvested in March/April. Con-sumption is from May to October, withthe maximum amount recorded in themonths of June and July.

The annual distribution of cases wasfound to be relatively constant except insituations of acute food shortage andfamine when epidemics of lathyrismhave been recorded. One such epidemicoccurred in the Dembia and Fogerasubdistricts in 1976/77 (11).

The clinical setting of the paralysiswas recalled in only 30% of thepatients, who remembered theprodromal symptoms of rheumatic painand tightness at the waist. Painfulspasms of the leg muscles, the so-calledmyospasm, was rarely recalled. Themajority experienced stiffness andheaviness of the legs at and around thetime of the paralysis of the legs, whichoccurred in a sub-acute manner in 65%

19

of the cases and acute (occurring all ofa sudden) in 25 %. The sub-acute onsetrefers to the presence of prodromalsymptoms before the manifestation ofthe paralysis and gradually increasingstiffness of legs over a month or more.Concerning disabilities, 52.2% and41.5% belonged to stage 1 and 2 respe-ctively. Stage 3 and 4 together formedonly 6.3%. Sixty-five percent of thefemales had the mild stage 1 disability(Table 2).

The dietary interviews revealed thatgrass-pea was commonly consumed assnacks in the boiled "nifro" and roasted"kollo" forms. The flour form was usedto prepare the Ethiopian sauce or gravycalled 'shiro wot'. In a typical grass-pea-producing village, this was thepredominant type of sauce consumedwith the injera, a pancake type ofleavened bread prepared from finger-millet, sorghum or rarely, from barley,wheat and tef (Eragrostis eff), depend-ing on the region and socio-economicalstanding of the household. Tef andwheat are generally expensive cerealsthat require good soil and a great dealof agricultural input. The bread (kitta)form of grass-pea is consumed onlyduring acute food shortages, and insome of the villages that were repeated-ly affected by flooding.

When interviewed about the con-nection of grass-pea and lathyrism, themajority of the victims were not clearthat the onset and severity of the paral-ysis had a direct relationship to theamount and period of grass-pea con-sumption. The belief that the paralysiswas precipitated or accentuated by

exposure to vapour or smoke fromgrass-pea being boiled or fried was verycommon. Walking in a grass-pea field,as well as drinking milk together withthe pulse, was thought to be dangerous.The interviews further revealed that theafflicted peasants continued to consumegrass-pea despite a clear understandingof its paralysing properties. When askedwhy they persisted in eating the "poi-sonous" pulse, they replied that therewas no alternative.

The clinical signs of the spasticparaparesis were uniformly similar tothe classical descriptions of neurolathy-ism. The victims manifest varyingdegrees of disability which cause diffi-culties with mobility and hinder theiroccupation as farmers. The commonprecipitating factors identified by thesesurveys were heavy physical labour,febrile illness and diarrhoea.

DISCUSSION

Ethiopia has been undergoing seriouspolitical turmoil for the last twentyyears. Because of military conflicts,many areas of the north, and later thecentral regions of the country, wereinaccessible to civilians. Ourepidemiological surveys were thereforehampered from operating in some of theadministrative regions where lathyrismwas known to be prevalent. Neverthe-less, the results now reported representthe best obtainable data on the lathyrismproblems of the major grass-pea grow-ing regions of the country.

Lathyrism is a real and serious prob-lem which has not received the attention

20

it deserves, either from the Ethiopianpeople or from its health authorities.The disease continues to claim moreand more victims in direct proportion tothe increasing cultivation and consump-tion of grass-pea. Those that havesuccumbed to the disease remain perma-nently disabled and are without oppor-tunities for social and physical rehabili-tation. Since young men are relativelycommonly affected there is a major lossof critical manpower for peasant fam-ilies.

The high prevalence rates oflathyrism in the Lake Tana basin (Table1) are very strongly influenced byseveral factors. First, the area is repeat-edly flooded during the rainy seasons(June to October), when cereals failbecause of waterlogging. Second, thearea is populated by very poor peasantswho live on subsistence farming, usingprimitive agricultural practices such asoxen ploughing. Their diet is cerealbased, with legumes being used forpreparation of snacks and sauces. Therewas hardly any intake of milk, proteinor vegetables (19).

In the 1988/89 re-survey of thedistricts of Dembia and Fogera ofnorth-west Ethiopia, a region that hadbeen affected by an epidemic in 1977(11), we found that the prevalence oflathyrism in some villages was as highas 3%. This was most alarming whenwe consider that the mean annual inci-dence of the disease in the same regionwas 0.6% (12). Although the preva-lence of the disease in other regionsmay be lower, the disease was never-theless found to be widespread, with its

maximal concentrations in northwestEthiopia, particularly in the administra-tive regions of West and East Gojamand South Gonder.

The high prevalence figures corre-spond to the cultivation and productionof grass-pea. The high BOAA levels ofgrass-pea from these regions may alsohave promoted disease outbreaks (Table1). Unfortunately, for reasons men-tioned earlier, we have been unable tosurvey the administrative regions ofNorth and South Welo and NorthShewa, where grass-pea production isrecorded to be high.

The characteristics of the affectedpopulations, their consumption of grass-pea, the onset and the clinical manifes-tations of the disease are very similar tothose in the classical reports from India(2-4, 19-20) and Bangladesh (21).

As shown in Fig. 2, the prevalenceof the disease in females is low andtheir degree of infirmity is less severe.This is similar to the results ofepidemiological studies in India (2,4). Itis not because women consume less, asActon suggested (2); our dietaryenquiry confirms that their consumptionis not less than that of men. The reasonfor the relative protection of females,particularly of child-bearing age, maybe hormonal (oestrogen), as proposedby Dwivedi and Prasad (4).

In India, legislation to ban the pro-duction and sale of grass-pea hasproved to be difficult to implement(22). Whereas in India serious attemptshave been made to tackle the problemof lathyrism, in Ethiopia very littleattention has been given to the preven-

tion of this crippling disease. Withnatural and man-made calamities socommonly encountered in Ethiopia, theproduction of the hardy pulse is increas-ing, and is bound to increase furtherparticularly in drought-prone and flood-prone areas. For instance, in 1988/89the annual production of grass-pea inthe country, from 37,800 hectares ofcultivated land, was estimated to be410,000 quintals (18).

Lathyrism is a very serious socialproblem because it cripples the youngand productive age groups. Diseaseprevention is an important public healthresponse. Acton, in his meticulouslyexecuted investigation of lathyrism inNorth Rewah, India (2) proposedmeasures that are both appropriate andfeasible in the Ethiopian situation:1. The public should be educated

about the toxicity of grass-pea,and instructed in ways of process-ing it to help reduce the toxicity.It is now well documented thatsoaking or boiling and discardingthe excess water reduces the toxicBOAA in the seed (23).

2. During poor harvest and faminesituations, communities in lathyrusgrowing areas should get externalrelief supplies free, or cerealsshould be made available to themat affordable prices.

Recently, interest and hopes havebeen raised by a more effective and far-reaching approach to solving the

21

lathyrism problem. This will involvethe development of low BOAA varietiesof grass-pea. It is interesting to observethat Acton, even in 1922, had alludedto this, suggesting that "the small-grained kesari is less toxic than thelarge grained Bhagalpur dal, so thatevery effort should be made to growsufficient quantity of indigenous grainto meet local requirements" (2).

There are now very active breedingprogrammes in a number of countries,including Ethiopia (14), to develop anddisseminate low-BOAA varieties ofgrass-pea. This is part of an interna-tional multidisciplinary collaborativeeffort to improve L. sativus and eradi-cate lathyrism (13).

ACKNOWLEDGEMENTS

These studies were made possible bythe Third World Medical ResearchFoundation (TWMRF) with a grantfrom Band-Aid. Dr. Aregay Waktola,Principal Investigator of the LathyrismProject in Ethiopia, Valeria Palmer ofTWMRF and Penny Jenden of Band-Aid are thanked for their interest andsupport. We are grateful to Ato NoadKebede and Dr. Birhanu Abegaz Mollaof the Chemistry Department, AddisAbeba University for the BOAA assayof grass-pea samples. We would alsolike to thank Ato Getachew GebreMariam who was instrumental in thesupervision of the field surveys.

22

TABLE 1. The prevalence of lathyrism in the surveyed areas comparedwith the Lathyrus sativus production and contentof beta-N-oxalylamino-L-alanine acid (BOAA)

of dry grass-pea samples.

Administrative region

North GonderDembia

South GonderFogeraSimadaKemkemDeraEste

West GojamYilmana DeinsaBahr Dar ZuriaMechaDega DamotJabir QuaritAchefer

East GojamMotaBichenaEnesieDejenEnarj-Enawga

West ShewaBecho

South ShewaButajira

Cases/pop.surveyed

999/160,000

767/123,500175/46,028113/62,927

169/174,90461/75,295

401/53,389148/44,43333/33,73915/20,91917/42,3761/14,162

29/14,54629/17,7159/16,549

14/30,9101/8,440

45/19,620

0/60,820

Prevalenceper 1,000

6.2

6.23.81.81.00.8

7.53.31.00,70.40.1

2.01.60.50.50.1

2.3

L. sativus prod*

area '000 ¡ha

1.4

11.4

8.1

13.1

3.5

0.28(0.5)

prod. '000qt.

10.3

69.1

77.8

102.1

31.6

1.8(4.9)

BOAA ingrass-pea

samples(mg/100gm)

0.44Î.85

0.5-0.96

0.45-0.85

0.38-0.78

0.2-0.50

* Production figure in hectares (ha.) and quintals (qt.) are from Agricultural SampleSurvey 1988/89; Central Statistics Office (CSO), 1990.

23

TABLE 2. Degree of disability by sex

Stage

1. No stick2. One stick3. Two sticks4. Crawler

Total

Males

No.

621607

5436

1,318

%

47.146.14.12.7

100.0

Females

No.

3391571510

521

%

65.130.12.91.9

100.0

Total

No.

9607646946

1,839

%

52.241.5

3.82.5

100.0

REFERENCES

1. Sleeman WH. Rambles and Recollections of an Indian Official. Vol 1. London:Hatchard and Sons, 1844.

2. Acton HW. An investigation into the causation of lathyrism in man. Ind MedGaz 1922;57:241-7.

3. Ganpathy KT, Dwivedi MP. Studies on clinical epidemiology of lathyrism.Lathyrism Enquiry Field Unit. Indian Council of Medical Research, GhandiMemorial Hospital, Rewa, Madhya Pradesh, 1961.

4. Dwivedi MP, Prasad VG. An epidemiological study of lathyrism in the districtof Rewa, Madhya Pradesh. Ind J Med Res 1964;52:81-114.

5. Rao SLN, Malathi K, Sharma PS. Lathyrism. World Rev Nutr Diet 1964; 10:214-38.

6. Spencer PS, Schaumburg HH. Lathyrism: a neurotoxic disease. Neurobehavio-ral toxicology and teratology 1983;5:625-9.

7. Dwivedi MP. Epidemiological aspects of lathyrism in India - a changingscenario. In? Spencer PS (ed). Grass-pea: Threat and Promise. New York: ThirdWorld Medical Research Foundation, 1989:1-26.

8. Hague A, Mannan MA. The problems of lathyrism in Bangladesh. In: SpencerPS (ed). Grass-pea: Threat and Promise. New York: Third World MedicalResearch Foundation, 1989:27-35.

9. Tekle Haimanot R. Lathyrism in Ethiopia. In: Spencer PS (ed). Grass-pea:Threat and Promise. New York: Third World Medical Research Foundation,1989:36-40.

10. Effo-Luzzi G. Malatia da "sebere" e lathyrismo in Eritrea. Bulletino dellasocieta Italiana di medicina e igiene tropicale 1947;7:483-93.

11. Gebre-Ab T, Wolde-Gabriel Z, Maffi M, Ahmed Z, Ayele TM, Fanta H.

24

Neurolathyrism - a review and a report of an epidemic. Ethiop Med J

12. Tekle-Haimanot R, Kidane Y, Wuhib E, Kalissa A, Alemu T, Zein AZ,Spencer PS. Lathyrism in rural northwestern Ethiopia: a highly prevalentneurotoxic disorder. Int J Epidemiol 1990; 19:664-72.

13. Palmer VS, Kaul AK, Spencer PS. International network for the improvementof Lathyrus sativus and the eradication of lathyrism (INILSEL): a TWMRFinitiative, In: Spencer PS (ed). Grass-pea: Threat and Promise. New York:Third World Medical Research Foundation, 1989;218-33.

14. Araya W, Dibabe A, Hundie B, Ensermu R, Haile W, Ashagrie Y. Grass-pea(Lathyrus sativus) research and its production potential in Ethiopia. (A paperpresented at the 3rd Triennial Colloquium of INILSEL, Dhaka, Bangladesh, 30November - 3 December 1991).

15. Tekle-Haimanto R, Abebe M, Gebre-Mariam A, Forsgren L, Holmgren G,Heijbel J, Ekstedt J. Community-based study of neurological disorders in ruralcentral Ethiopia. Neurepidemiology 1990;9:263-77.

16. Rao SLN. A sensitive and specific calorimetric method for the determinationof L.B. diaminoproprionic acid and the Lathyrus sativus neurotoxin. AnalBiochem 1978;864:386-95.

17. Abegaz B, Kebede N, Asmelash S. HPLC analysis of B-N-Oxalylamino-L-alanine, the neurotoxin in the legume Lathyrus sativus. In: Spencer PS (ed).Grass-pea: Threat and Promise. New York: Third World Medical ResearchFoundation, 1989; 128-32.

18. Ethiopia. Central Statistics Office. 1990. Agricuture sample survey 1988/1989.Result on area, production and yield of major crops by sector and season.Statistical Bulletin no. 79 ONCCP, Addis Abeba.

19. Kulkarni SW, Attal HC, Choubey BS. An epidemiologic study of lathyrism inAmgaon Block, Bahandra district. Ind Med J Res 1977;66:602-10.

20. Attal HC, Kulkarni SW, Choubey BS, Palkar ND, Deotale PG. A field studyof lathyrism - some clinical aspects. Ind J Med Res 1978;67:608-15.

21. Ludolph AC, Hugon J, Dwivedi MP, Schaumburg HH, Spencer PS. Studieson the aetiology and pathogenesis of motor neuron diseases. 1. Lathyrism:clinical findings in established cases. Brain 1987; 110:149-65.

22. Dwivedi MP, Singh SP. The lathyrism problem: Current status and newdimension. In: Scientific Report, Nutrition Foundation of India. 1984; 19-33.

23. Mohan VS, Nagarajan V, Gopalan C. Simple practical procedures for theremoval of toxic factors in Laythyrus sativus (Khesari dal). Ind J Med Res1966;54:410-9.

(Accepted 2 November 1992)