Clinical manifestations

Presentation and course

Lathyrism is a form of irreversible, non-progressive spastic paraparesis associated with poorly understood degenerative changes in the spinal cord. Often, the patient awakens one morning to find lower-limb weakness, causing him or her to fall. Severe cramps often accompany or herald the weakness, which can be precipitated by sudden physical exertion or exposure to inclement weather. The acute phase lasts for about a month, and residual symptoms persist. In a few patients, the disease advances subacutely over a month, followed by slow progression. Some patients experience a prolonged slow progression (29; 48). Neurologic examination often reveals a pyramidal pattern of motor weakness combined with marked hypertonia in the thigh extensors and adductors and in the gastrocnemius muscles so that the more severely affected patients walk on the balls of their feet with a lurching scissoring gait. Extensor plantar responses are uniformly present, the knee and ankle tendon reflexes are exaggerated, and clonus is often present. Hoffmann-Trömner sign and exaggerated biceps or triceps tendon jerks are also found in the most severely affected patients. Sensory and bladder manifestations are infrequent (29). Animals fed peas from Lathyrus species develop skeletal deformities and aortic aneurysms rather than neurologic disorders. This animal disease primarily affects collagen tissue and is called osteolathyrism (05). The occurrence of osteolathyrism in humans is poorly documented. Some patients develop bone pain and disfigurement of bones, including vertebrae and pelvis.

Konzo is a neurologic disorder of abrupt onset with selective upper motor neuron damage, characterized by an irreversible, non-progressive, and symmetrical spastic paraparesis or tetraparesis (37; 48). The World Health Organization provided the following criteria for konzo: (1) abrupt onset (less than 1 week) of non-progressive lower limb weakness in a healthy person; (2) symmetric spasticity; and (3) bilaterally increased lower limb reflexes in the absence of signs of spine disease (56). Children and lactating women are the most vulnerable (32; 55). In the majority of patients, the symptoms evolve within a few days. The initial symptoms include cramps, a heavy feeling or weakness in the legs, difficulty standing and walking, and a tendency to fall down. Occasionally, there is lower back pain, blurred vision, speech difficulties, and paresthesias of the legs, but these symptoms are often transient and disappear within a month. In most patients, during the first 2 days, the disease is serious enough to force them to bed. Later on, there may be a partial improvement. Finally, the affected person develops a stable spastic paraparesis, which persists for the remainder of life. After onset, the neurologic signs remain constant or improve minimally if no further cyanide is ingested, unlike, for example, human T-cell lymphotropic virus -I infection, in which further deterioration takes place (41; 21). In addition to motor deficits, konzo-affected children also have cognitive impairment. The determinants of cognitive impairment in these children include female gender and low serum albumin (11). Half the patients with konzo have an optic neuropathy with decreased visual acuity, alterations of the visual fields, and atrophy of the temporal part of the retinal nerve fiber layer linked with temporal pallor of the optic disc. About half of the konzo patients also have abnormal visual evoked potentials, with both delayed latency and reduced amplitude. A few of these patients have an ocular motor disturbance, leading to pendular nystagmus (34). One study demonstrated significant motor and cognitive impairment in boys with konzo. Motor and cognitive testing were done by Bruininks-Oseretsky Test of Motor Proficiency and Kaufman Assessment Battery for Children, respectively (08). Follow-ups were at 2 and 4 years of age. Kaufman Assessment Battery for Children scores of boys with konzo were significantly lower at 2 years of follow-up than the boys who do not have konzo. For boys, increases in urinary thiocyanate concentration were significantly associated with reductions in Bruininks-Oseretsky Test of Motor Proficiency but not for cognitive decline (08; 35).e

Tropical ataxic neuropathy typically occurs in endemic form and has a slowly progressive course. In contrast to konzo, tropical ataxic neuropathy is more common in persons older than 40 years. Tropical ataxic neuropathy is a neurologic syndrome of sensory polyneuropathy, gait ataxia, optic atrophy, and neurosensory deafness. The presence of two of the four features (sensory polyneuropathy, gait ataxia, optic atrophy, and neurosensory deafness) is considered enough to establish the diagnosis. This syndrome typically begins with dysesthesias in the feet, followed by unsteadiness of the gait. There is often a loss of vibration and proprioceptive sensations. A Romberg test is often positive. Reflexes in lower limbs are often diminished or absent. Plantar responses are flexor. Spastic paraparesis is present in only 20% of patients (41; 40; 48). A 2008 study reported, for the first time, evidence for the occurrence of tropical ataxic neuropathy in south India (31). The authors reviewed the clinical and laboratory profiles of 40 tropical ataxic neuropathy cases. The notable demographic characteristics of cases in India included female preponderance, peak age at onset in the thirties, rural residence, and poor socioeconomic status. The diet in the majority comprised a large amount of tapioca, which was low in protein. In addition to sensory peripheral neuropathy, 90% had decreased hearing, 50% had decreased vision, and 25% had spasticity involving the lower extremities. Compared to the controls, the serum, urine, and sural nerve thiocyanate levels were significantly elevated in the patients. With cessation of cassava intake and better nutrition, improvement of the neurologic disability occurred in the majority.

Prognosis and complications

Acton graded the disability of lathyrism into four stages according to the support required in walking. Most patients continue to walk slowly without a stick (no-stick stage); some patients need one or two sticks (one-stick stage or two-stick stage); on further progression, patients can move around on both hands and knees (crawler stage) (01).

In a follow-up study, nine patients with konzo were reexamined after 14 years. This long-term follow-up showed that the neurologic signs in patients with konzo remained constant. Four severely affected patients showed functional improvement (13).

A study showed that endemic ataxic polyneuropathy led to a decrease in the survival of affected patients. In this study, a total of 204 cases and 5766 controls (4000 controls in the endemic community and 1766 controls in the nonendemic community) were followed for 25 months. A total of 153 subjects died during follow-up, including 24 endemic ataxic polyneuropathy cases, 115 controls in the endemic community, and 14 controls in the nonendemic community. Relative risks of death, adjusted for age and gender, were 4.5 for cases (P < 0.0001) but 2.6 for controls living in the endemic community (P = 0.001). Authors were of the opinion that the finding of lower risk of death in the community with higher exposure to cyanide from cassava foods indicated that mortality of endemic ataxic polyneuropathy was not associated with exposure to cyanide from cassava foods (38).

Biological basis

Etiology and pathogenesis

The Lathyrus sativus plant is still cultivated in India, Bangladesh, Nepal, Pakistan, Iran, the Middle East, southern Europe, Ethiopia, and some places in South America. Seeds of this legume plant, which is grown in arid and semi-arid tropical regions, contain beta-N-oxalyl amino-L-alanine (synonym beta-N-oxalyl-L-alpha,beta-diaminopropionic acid). Beta-N-oxalyl amino-L-alanine is a neurotoxin that, when consumed, causes lathyrism. The plant L sativus flourishes in conditions of the most devastating flood and drought, often when no other food crop survives. The product of this especially hardy plant, the chickling pea, historically has served as an inexpensive survival food for the poor population of poor and developing countries (30; 44). An epidemiological study from Ethiopia suggested that in areas with high neurolathyrism prevalence, prevalence is significantly influenced by age, gender, marital status, family size, grass pea production levels, and dietary habits. The study suggests that exclusive consumption of grass pea, particularly immature seeds and in high ratios compared to other cereals, poses a considerable risk for developing neurolathyrism, emphasizing the need for diversified diets and awareness to mitigate this risk (07).

The epidemics of konzo occur in the dry season among very poor rural populations, whose diets for the weeks and months before onset consist almost exclusively of bitter cassava roots. The root of the cassava (Manihot esculenta Crantz) shrub is an important staple food for millions of people in tropical and subtropical countries. Epidemic outbreaks in subtropical and tropical regions are associated with drought-provoked agricultural and social crises, rendering populations dependent on a diet of insufficiently processed bitter cassava. The shortcuts in processing allow large amounts of cyanogens to remain in the cassava consumed; hence, there is a high dietary cyanide exposure. The drought increases the natural occurrence of cyanogenic glucosides in the cassava roots, and the processing procedure normally applied to remove cyanide before consumption is shortened because of food shortage (53; 30; 37). Children with konzo had low levels of selenium, copper, and zinc compared to controls. Motor deficits associated with konzo may be due to a lack of trace elements-induced cyanide-scavenging (10). The diet of konzo-affected patients is usually deficient in protein content (55). A systematic review by Baguma and colleagues revealed that cassava-derived cyanide poisoning and protein malnutrition were two predominant risk factors in the pathogenesis of konzo. Authors further noted that thiocyanate, the biomarker of choice for cyanide exposure, does not accurately estimate the actual level of cyanide poisoning in konzo patients (03).

Tropical ataxic neuropathy is also encountered frequently in malnourished populations. Tropical ataxic neuropathy is frequently observed in populations that use large quantities of cassava in their diets for long periods. Some observations, however, do not suggest that cyanide is the cause of endemic ataxic polyneuropathy (39).

Beta-N-oxalyl amino-L-alanine (synonym beta-N-oxalyl-L-alpha,beta-diaminopropionic acid) is an excitatory amino acid and acts as an agonist at the alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)-kainate receptors located on specific neurons and their dendrites. One of the primary effects of beta-N-oxalyl amino-L-alanine toxicity is the selective inhibition of mitochondrial complex-1 in the motor cortex and lumbar spinal cord (47). Streifler examined the central nervous system of a 67-year-old man who showed symptoms of lathyrism after being imprisoned at the age of 35 and fed a diet of chickling peas, Lathyrus sativus. The most obvious changes were a loss of axons and myelin in the pyramidal tract in the lumbar spinal cord. These alterations correlated with spastic paraparesis. The longest pyramidal neurons are more severely affected (44). Another change at the same level was a mild degree of degeneration of anterior horn cells. A few examples of senile plaques and neurofibrillary tangles were observed in Ammon's horn (50; 17). A report of a patient who had a disease 32 years before death revealed some changes in anterior horn cells with swellings, Hirano bodies, and dispersion of Nissl bodies (20). In a rat model of lathyrism, it was observed that the number and size of motor neurons were decreased only in the lumbar and sacral cord segments (26). In addition, extensive but transient hemorrhage was noted in the ventral spinal cord parenchyma of the rat. Vascular endothelial growth factor receptor levels were significantly lowered in the lumbosacral spinal cord of the paraparetic rats compared with their controls, suggesting a failure of the vascular endothelial growth factor system to protect neurons against beta-N-oxalyl-L-alpha,beta-diaminopropionic acid toxicity. Some studies have indicated that nutritional deficiencies of sulfur amino acid methionine and cysteine aggravate the neurotoxicity of beta-N-oxalyl amino-L-alanine (57).

A number of pathogenic mechanisms for the development of konzo and tropical ataxic neuropathy among inherently susceptible persons have been proposed. The metabolites of linamarin (α-Hydroxyisobutyronitrile β-D-glucopyranoside, dominant cassava cyanogen), cyanide (mitochondrial toxin), thiocyanate (AMPA chaotropic agent), and cyanate (protein carbamylating agent) are considered responsible for the pathogenesis of konzo (23).

The bitter varieties of cassava have a relatively high content of cyanide. Many plants liberate hydrogen cyanide from cyanogenic glycosides, a phenomenon called cyanogenesis, as a defense mechanism against animals and insects. Cassava is by far the most important human food source that uses cyanide as a defense mechanism (37). Cyanide is metabolized to several compounds, such as thiocyanate, cyanate, formate, and 2-iminothiazolidine-4-carboxylic acid, but the major metabolite is thiocyanate. The body uses sulfur-containing amino acids to render cyanide harmless. Thiols, which are sources of sulfur, are required to detoxify cyanide to thiocyanate. It has been hypothesized that the detoxification of cyanide to thiocyanate may be impaired in cases of tropical ataxic neuropathy because of the nutritional deficiency of thiols. If the diet is deficient in sulfur, cyanide will be converted to cyanate, which induces neurodegenerative disease in both animals and humans. Mainly children and ill people, often malnourished (and with kwashiorkor), are not able to detoxify the cyanide sufficiently, consequently developing symptoms of upper motor neuron damage (14; 51). Thiocyanate selectively promotes the action of excitatory amino acid glutamate at AMPA receptors. Some investigators have even suggested that cyanide is a potent inhibitor of mitochondrial enzyme cytochrome-c-oxidase. It has been suggested that cassava contains a chemical known as cyanoglucoside linamarin (alpha-hydroxyisobutyronitrile-beta-d-glucopyranoside). Linamarin is enzymatically converted to cyanide by bacteria in the intestine, and this is absorbed into the blood and then damages neural cells (49). It has been postulated that in Konzo patients, a severe exacerbation of thiamine deficiency results from the inactivation of thiamine that occurs when the sulfur in thiamine is utilized for the detoxification of cyanide consumed in improperly processed bitter cassava (02).

In patients of konzo, the neuronal cells that are most likely affected are Betz cells in the motor cortex. There are no autopsy reports on konzo available. In two original cases, described by Trolli, spinal cords were reported normal. Autopsies performed in ex-prisoners of war suffering from tropical ataxic neuropathy revealed demyelination of the posterior column (particularly of the fasciculus gracilis) and the optic nerves. Nerve biopsy revealed nonspecific patchy demyelination with variable pericapillary cellular reaction and perineural fibrosis (52).

A review presented a unifying hypothesis on the causative agents of lathyrism, konzo, and tropical ataxic neuropathy (28). This review suggested that these diseases may be caused by a group of cyano-containing compounds called nitriles. Direct neurotoxic actions of nitriles are not mediated by systemic cyanide release. Both cassava and Lathyrus contain nitriles, and other unidentified nitriles can be generated during food processing or in the human body. Available data indicate that several small nitriles cause a variety of neurotoxic effects. In experimental animals, 3,3'-iminodipropionitrile, allylnitrile and cis-crotononitrile cause sensory toxicity, whereas hexadienenitrile and trans-crotononitrile induce selective neuronal degeneration in discrete brain regions (28). 3,3'-iminodipropionitrile also induces a neurofilamentous axonopathy, and dimethylaminopropionitrile is known to cause autonomic neurotoxicity in both humans and rodents.

A study suggested a causal association between konzo and the gut microbiome (09). Bramble and colleagues utilized a shotgun metagenomic sequencing technique in 180 individuals in the Democratic Republic of the Congo. Samples were drawn from different geographical regions of Congo, including participants from konzo-prone regions. The gut flora of individuals varied from region to region. Despite similarities in levels of nutritional deficiency, lifestyles, and diets high in cyanogenic cassava, the relative abundance of gut flora in unaffected children from these two konzo-prone areas was significantly different. The authors concluded that the unique gut flora potentially modulates the susceptibility to konzo in konzo-prone regions of the Democratic Republic of the Congo.

Epidemiology

Once prevalent throughout Europe, North Africa, the Middle East, and parts of the Far East, lathyrism is restricted to India, Bangladesh, and Ethiopia. A survey was conducted in an Ethiopian district (a known endemic area for lathyrism), and disease information was obtained for 2987 family members after interviewing 589 randomly selected heads of households. Neurolathyrism patients were detected in 56 households (9.5%, prevalence rate 2.38%). The mean number of affected family members per household was 1.27. The median age at onset of paralysis was 11 years (range 3 to 44 years) (15). Two northwestern districts of Bangladesh with a population of 629,752 were surveyed from June 1991 to March 1993 to detect and follow up with lathyrism patients. Out of 2567 neurologic patients, 882 were diagnosed as having lathyrism, giving a prevalence rate of 14 per 10,000. This prevalence was higher among young males: only 12.9% of the patients were female, and only 19.3% were over 30 years of age at onset of the disease (19). A door-to-door epidemiological survey for the disease was carried out in the major areas of northwest and central Ethiopia where Lathyrus sativus was grown. The survey involved a population of 1,011,272. A total of 3026 affected persons were identified. Variable prevalence rates in different regions, ranging from one per 10,000 to 7.5 per 1000 were observed. The male-to-female ratio of cases was 2.6 to 1; the females exhibited a milder form of the disease (18). An epidemiological study performed in Northeast Ethiopia noted that overall prevalence of lathyrism among 2307 inhabitants was 5.5% (22). Men were more frequently affected than women.

In areas of India where a high consumption of Lathyrus sativus was previously reported, increased awareness of the disease has led to a marked reduction in lathyrism cases. A total of 17,755 (13,129 rural and 4626 urban) inhabitants from 151 villages and 60 wards from urban areas were clinically examined, and only nine old cases of lathyrism were found. Total consumption of Lathyrus sativus had also drastically decreased (25).

The majority of cases of konzo occur during epidemics. In the last decade, several epidemics of konzo have been reported among the African rural populations, including Mozambique, Tanzania, Zaire, and the Central African Republic. Konzo affects mainly women and children. Breastfed children and adults over 50 years of age often remain spared from konzo. A strong familial clustering during epidemics has been noted. The prevalence rate of konzo in affected populations ranges from 1 to 30 per 1000. Two large epidemics have been reported, each of more than 1000 cases. The first was in the Bandundu region in Congo (1936 to 1937), and the second was in the Nampula province of Mozambique (1981). Small outbreaks have been reported from other places. Sporadic cases of konzo also occur. Most cases of konzo occur in the dry season, chiefly during a long drought. Cliff and colleagues reported an epidemic of konzo in which 384 patients were treated in rehabilitation centers. The prevalence rate in a badly affected area was 30 per 1000. Most patients were children over the age of 3 years and women (14). In a report of Médecins Sans Frontières’s demographic data, history, and clinical presentation were recorded for each konzo patient among 60,000 refugees from the Central African Republic who sought refuge in villages in eastern Cameroon (between 2005 and 2007) (12). A total of 469 patients were diagnosed with konzo. Children and women of reproductive age were dominantly affected. Most patients developed symptoms in a seasonal pattern, with most of the cases occurring during the dry winter season. Most of the patients complained about walking difficulties and weight loss and had exaggerated lower limb reflexes and muscle wasting.

The prevalence of tropical ataxic neuropathy in some parts of Africa ranges from 29 to 34 cases per 1000 population. A study determined the prevalence of tropical ataxic neuropathy in southwestern Nigeria. A total of 4583 inhabitants were registered in the census. Of these, 3428 subjects aged 10 years and older were screened. The diagnosis of tropical ataxic neuropathy was made in 206 of 323 subjects who screened positive for tropical ataxic neuropathy. The prevalence of tropical ataxic neuropathy was 6% (4% in males and 7.7% in females). The highest age-specific prevalence was 24% in the 60 to 69 years age group in women (40).

In an epidemiological study in Nigeria, 2392 individuals aged 18 years or older (a country with high consumption of cassava-based meals) were screened for tropical ataxic neuropathy. In this study, 39 cases of tropical ataxic neuropathy were diagnosed with a rough prevalence rate of 16.3 out of 1000. Distal sensory polyneuropathy was the most common clinical manifestation (43).

Prevention

Consumption of pure grass pea, especially in the green unripe and boiled forms, should be avoided. Communities at risk of neurolathyrism during famine crises should be encouraged to combine and use grass pea with cereals before the household cereal stock is fully depleted (16). Detoxification of grass pea seeds requires the removal of the husk and boiling of the seeds for several hours in hot water. The supernatant contains about 70% to 80% of the neurotoxin originally present in the seeds. This water is discarded. The scarcity of water during a drought increases the risk that the grass pea seeds will not be properly detoxified (05). Improved cassava processing will help reduce the incidence of both konzo and tropical ataxic neuropathy. It is advised to broaden the diet of people through the introduction of other staples, cereals, and fruits. Detoxification of cassava leaves can be achieved through a combination of simple processing methods like pounding and boiling. An experimental study on rats noted that Ginkgo biloba extracts can prevent motor changes associated with cassava juice consumption.

A ‘‘wetting method’’ has been reported to eradicate or markedly reduce konzo outbreaks in the Democratic Republic of Congo (04).

Differential diagnosis

Confusing conditions

The diagnosis of lathyrism depends on the clinical features. The diagnostic points include consumption of large amounts of Lathyrus sativus in the weeks before the disturbed gait; markedly spastic paraparesis; normal cranial nerves and cerebellar and sensory examinations; no other obvious reason for the paraparesis. Several other non-compressive myelopathies, like primary lateral sclerosis, nutritional myelopathies, hereditary spastic paraplegia, human immunodeficiency virus-myelopathy, and multiple sclerosis, may be mistaken for lathyrism. Even early spinal cord compression may pose a diagnostic problem.

The clinical manifestation of human T-cell leukemia/lymphoma virus type-1/tropical spastic paraplegia is similar to konzo and neurolathyrism, and this neurologic disease has been considered a differential diagnosis of konzo and neurolathyrism. The three diseases occur in similar socioeconomic settings. As in konzo and neurolathyrism, environmental factors and geographic locations seem to be important in the etiology of tropical spastic paraparesis. Many cases of tropical spastic paraparesis are not linked to human T-cell leukemia/lymphoma virus type-1 infection, and some experts consider konzo and neurolathyrism as variants of tropical spastic paraparesis (36).

Diagnostic workup

There is no specific diagnostic test for lathyrism; diagnosis relies only on clinical evaluation.

In patients with konzo and tropical ataxic neuropathy, urinary concentrations of thiocyanate and linamarin are elevated in proportion to cassava consumption. The patients are human T-cell lymphotropic virus-1 negative. Two Tanzanian patients with konzo, who were severely disabled by a non-progressive spastic paraparesis, were subjected to neuroimaging and electrophysiological studies. MRI of the brain and spinal cord were normal. However, motor-evoked potentials on magnetic brain stimulation showed a complete absence of response in the legs and even slightly affected upper limbs. Other neurophysiological investigations, including electroencephalography, were largely normal (54). Electrophysiological evaluation of tropical ataxic neuropathy often reveals normal motor nerve conduction findings. However, there may be a slowing of sensory nerve conduction velocities. Delayed P100 amplitude of the visual evoked potential suggests demyelination of the optic nerves (39).

Management

In cases of lathyrism, continued consumption of Lathyrus sativus as a staple food should be avoided. Disability in lathyrism is permanent. Spasticity is responsible for a large part of the disability. Usage of antispasticity drugs, such as baclofen and diazepam, is helpful. Patients can benefit from physiotherapy and surgical release of tendon contractures.

There is also no definite treatment available for konzo. A good diet containing high doses of vitamins and physical rehabilitation are recommended. Treatment with sodium thiosulfate (Na2S2O3), a cyanide antidote, did not provide encouraging results. Locally initiated rehabilitation programs have proven extremely useful for these patients. Likewise, there is no specific treatment for tropical ataxic neuropathy. It does not respond to vitamin supplements or an improved diet.