HAT is a life-threatening illness caused by infection with extracellular protozoan parasites that are transmitted by tsetse flies in sub-Saharan Africa. T. b. gambiense and T. b. rhodesiense are the two pathogenic subspecies affecting humans; their epidemiologic and clinical features largely differ.
The geographic range of HAT is restricted to sub-Saharan Africa in line with the distribution of its vector, the tsetse fly (Glossina species; Fig. 222-3). HAT due to T. b. gambiense is endemic in 24 countries of western and central Africa. Between 1999 and 2015, the number of reported cases fell by 90% (from 27,862 to 2733) as a result of successful control measures based on systematic screening of populations at risk, diagnostic confirmation, and treatment of infected individuals. During the same period, the number of reported cases of HAT due to T. b. rhodesiense fell by 89% (from 619 to 71) in the 13 disease-endemic countries of eastern and southeastern Africa. However, the ratio of reported to unreported cases remains uncertain for disease caused by both species. In 2015, most cases of T. b. gambiense HAT were reported by the Democratic Republic of the Congo (DRC; 86%), whereas Malawi and Uganda reported most of the cases caused by T. b. rhodesiense (42 and 39%, respectively). The geographic distributions of T. b. gambiense and T. b. rhodesiense do not overlap, but the two species are present in distinct regions of Uganda and the DRC. A roadmap for HAT elimination as a public health problem is being mapped out by the World Health Organization; the objective is to decrease the frequency of new cases to <1 per 10,000 people in endemic areas by 2020.
Areas at risk for human African trypanosomiasis, 2010–2014. (Courtesy of Jose Ramon Franco, MD.)
Humans are the predominant or exclusive reservoir of T. b. gambiense. Rare cases of vertical (in utero) or transfusional transmission have been reported, but almost all patients are infected by the bite of tsetse flies during their daily activities along or near rivers, where the flies live and reproduce. In contrast, T. b. rhodesiense causes zoonosis in a variety of wild and domesticated animals (e.g., antelopes and cattle, respectively), which act as reservoirs. Humans are infected by T. b. rhodesiense via tsetse bites in woodland savannah. Honey gatherers, game park rangers, poachers, and firewood collectors are particularly at risk. Imported cases of HAT are occasionally diagnosed among African immigrants and other travelers. While long-term travelers (>30 days) are at increased risk of T. b. gambiense HAT, most imported cases of T. b. rhodesiense HAT are seen in short-term travelers, typically following visits to game parks.
PATHOLOGY AND PATHOGENESIS
T. b. rhodesiense and T. b. gambiense, unlike other trypanosome species, can infect humans because they resist lytic factors in human serum—namely, apolipoprotein L-1. The serum resistance–associated protein is responsible for resistance in T. b. rhodesiense, whereas other mechanisms, notably involving the T. b. gambiense–specific glycoprotein (TgsGP) gene, are used by T. b. gambiense.
Trypanosomes are transmitted to humans by the tsetse bite, proliferate, and induce a local inflammatory reaction that is sometimes clinically apparent as a chancre. Trypanosomes then disseminate into the hematolymphatic system, with lymph nodes becoming enlarged after infiltration by mononuclear cells and lymphocytes. The degree of enlargement of the liver and spleen is usually mild to moderate, with infiltration by mononuclear cells as a prominent feature. Trypanosomes multiply in the blood, but their presence and density vary. This variation is mainly due to a cyclic immune-evasion process, whereby the parasite population can be decimated by the host’s immune response until the reemergence of offspring parasites that express a different variant surface glycoprotein to which the immune system is temporarily blind. Each trypanosome genome encodes a repertoire of ~1000 variant surface glycoproteins between which the parasites can switch genetically. Trypanosomes also multiply in extravascular tissues during the first stage of illness. The skin, skeletal muscles, serous membranes (peritoneum, pleurae, and pericardium), and heart can be involved, with interstitial infiltration of mononuclear cells and vasculitis evident on microscopic examination. Myocarditis and pericarditis with myocardial degeneration and interstitial hemorrhage are common features of T. b. rhodesiense infection.
The CNS is invaded weeks to months (T. b. rhodesiense) or months to years (T. b. gambiense) after initial infection. This invasion corresponds to the second stage of HAT, which is defined by the presence of trypanosomes or mononuclear cells in the cerebrospinal fluid (CSF). The white matter is predominantly affected, with perivascular proliferation of astrocytes, microglial cells, and Mott’s (morular) cells that contain IgM in intracellular vacuoles. The location of white-matter lesions in the brain correlates with the main neurologic clinical features. The cerebral cortex and neurons are spared until the terminal stages of illness. Because reversible inflammatory lesions predominate over the irreversible destruction of tissue, neuropsychiatric symptoms and signs resolve partially or completely during or after treatment of second-stage HAT.
APPROACH TO THE PATIENT Human African Trypanosomiasis
HAT is usually lethal in the absence of treatment, and treatment is simpler and safer during the first stage of illness. Therefore, early diagnosis is crucial; physicians should include HAT in the differential diagnosis of several clinical syndromes when a patient has traveled or lived in at-risk sub-Saharan African countries, and obtaining a thorough recent and remote travel history from the patient is a prerequisite for diagnosis. In particular, HAT due to T. b. gambiense should be suspected in patients with persistent and intermittent fever or headaches, progressive neuropsychiatric disorders, and biological signs of systemic inflammation, even if the last exposure occurred several years previously. HAT due to T. b. rhodesiense should be suspected in patients with an acute febrile illness and a recent exposure to tsetse flies in an eastern African country, especially if diagnostic tests for malaria are negative.
The clinical presentations of T. b. gambiense and T. b. rhodesiense HAT usually differ. T. b. gambiense HAT is a slowly evolving illness with a long incubation period (months to years) and a prolonged disease course. In contrast, T. b. rhodesiense HAT is an acute febrile illness with a short (<3-week) incubation period and a shorter (weeks to months) disease course. There are exceptions to this classical pattern. Acute forms of T. b. gambiense HAT have been reported, especially among travelers, and chronic forms of T. b. rhodesiense HAT occur in the southern range of its geographic distribution (e.g., Zambia and Malawi). Trypanotolerance—i.e., the long-term persistence of parasites without clinical features of disease—is increasingly being reported for T. b. gambiense. Concomitant HIV co-infection does not seem to predispose individuals to an increased risk of HAT, and the virus’s impact on the clinical presentation of HAT is not known.
The occurrence of trypanosomal chancre is reported in a sizeable proportion of travelers, but very rarely in patients living in endemic areas, where the nonpurulent, painful, and itchy nodule can easily be confused with the bite of another arthropod. The chancre spontaneously disappears in 1–3 weeks.
After an asymptomatic incubation period that usually lasts for weeks or months but occasionally lasts for years, patients may present with irregular and remittent fever, sometimes accompanied by fatigue, malaise, and myalgia. Fever is more frequent among travelers than among natives, but the absence of fever in no way rules out the disease. Circinate or serpiginous rashes, commonly called trypanids, can occur on the trunk and on proximal parts of the extremities. Trypanids are almost impossible to detect on dark skin and have been reported only in Caucasians. Pruritus is a common but nonspecific symptom that affects up to half of patients during the second stage. Painless edema of the face and extremities occasionally occurs during the first phase.
Enlarged lymph nodes—a classical sign of HAT—are detected in 38–85% patients at both disease stages. Cervical palpation is essential in patients with suspected HAT. The lateroposterior cervical group (Winterbottom sign) and the supraclavicular group are most commonly affected. Lymph nodes are movable, soft initially, harder later, and painless. A variable proportion of patients present with mild to moderate hepatomegaly and splenomegaly. Signs of myocarditis and pericarditis are occasionally detected by ECG and echocardiography but are usually clinically silent. Symptoms of HAT may mimic hypothyroidism or adrenal insufficiency, but thyroid and adrenal function tests yield normal results. Loss of libido, impotence, and amenorrhea, with decreased levels of testosterone and estradiol, are common in second-stage patients and are most likely caused by dysfunction of the hypothalamic–pituitary axis.
Most patients with second-stage illness have no or only mild specific neuropsychiatric symptoms and signs, which, when they develop, tend to do so late in the disease course. In contrast, some nonspecific features, such as headaches and mood and behavioral changes, are present in both disease stages but become more permanent and severe during the second stage. As mentioned earlier, HAT is commonly called “sleeping sickness” because of various sleep disturbances (daytime somnolence, nocturnal insomnia) that are more pronounced later in the second stage. Dysregulation of the daily sleep/wake cycle and fragmentation of sleeping patterns are characteristic. Depending on the area of the brain affected, various neurologic syndromes can also develop, including disorders that are pyramidal-related (e.g., motor weakness, rare instances of hemiplegia), extrapyramidal-related (e.g., rigidity, paratonia), and cerebellar-related (e.g., ataxia, abnormal gait). Fine tremor, resting myoclonus, and abnormal (athetoid or choreic) movements have also been reported. Mental disorder is a key feature of HAT and can easily be misdiagnosed as primary psychiatric illness. Common presentations are antisocial or aggressive behavior, mood disorders (e.g., irritability, indifference), apathy or hyperactivity, and depression or psychosis (e.g., delirium, hallucinations). In the final stage of illness, decreased consciousness, dementia, and sometimes epilepsy are present, leading to coma, bed sores, aspiration pneumonia, or other bacterial infections and ultimately to death.
The clinical presentation of T. b. rhodesiense HAT can be similar to that of T. b. gambiense HAT in areas (e.g., Zambia, Malawi) that characteristically harbor specific parasite genotypes and host factors. The typical acute form with an incubation period of <3 weeks occurs in the northern range of the disease’s distribution (e.g., Tanzania, Uganda) and in travelers. The initial trypanosomal chancre is clinically similar to that seen in T. b. gambiense HAT but is more common, especially among travelers.
Fever can be high and occurs in both first- and second-stage patients, often in association with headaches and with diffuse myalgia and arthralgia. Pruritus and edema of the face and legs can be present. Lymphadenopathies have been reported in variable proportions in both disease stages and predominately affect the submandibular, axillary, and inguinal regions. Mild to moderate hepatomegaly and splenomegaly are documented in a minority of patients. Myocarditis and pericarditis appear to influence clinical course and outcome, even though clinical features of cardiac failure or arrhythmia have not been prominent findings in large case series. In contrast, conduction abnormalities, with various degrees of atrioventricular block, have been reported in travelers. Sepsis-like features, with disseminated intravascular coagulation and multiple-organ failure, can occur in the terminal stage.
Neuropsychiatric symptoms and signs in T. b. rhodesiense HAT are reported with varying frequency but overall are similar to those described above for T. b. gambiense HAT. The notable exception in T. b. rhodesiense disease is a more rapid evolution toward coma and death.
The clinical and biological features of T. b. gambiense and T. b. rhodesiense HAT—anemia, thrombocytopenia, elevated levels of C-reactive protein and IgM—are not sufficiently specific and current drug regimens are not sufficiently simple and safe to allow the initiation of treatment solely on the basis of suspicion. Diagnostic confirmation is therefore mandatory in all patients.
The diagnosis of T. b. gambiense HAT is based on a three-step approach: screening, diagnostic confirmation, and staging.
Immunologic (serologic) methods constitute the preferred screening tool. The card agglutination test for trypanosomiasis (CATT) has been used in most endemic areas for several decades. The test reagent contains stained, freeze-dried trypanosomes of selected variable-antigen types. If specific antibodies are present in the patient’s blood or serum, agglutination can be seen with the naked eye. The sensitivity of the CATT on undiluted blood or serum is 69–100% (>90% in most studies), with some regional variation; its specificity is 84–99%. The CATT and associated equipment (e.g., a rotator) are manufactured and distributed by the Institute of Tropical Medicine in Antwerp, Belgium, but are not widely available outside endemic areas. In recent years, lateral flow tests have been developed and commercialized, first based on whole parasites and later on recombinant antigens. Their diagnostic performance appears similar to that of the CATT. Other serologic test formats (ELISA, immunofluorescence, indirect hemagglutination) are available in some reference laboratories in both endemic and nonendemic countries.
The microscopic observation of trypanosomes in the lymph, blood, or CSF confirms the diagnosis. Direct observation of motile trypanosomes on a wet preparation of lymph obtained by cervical lymph-node puncture is simple and cheap but has limited sensitivity (50–65% in most studies). Trypanosomes can be found in the blood but often occur at low densities. Therefore, stained thin and thick blood smears have very low sensitivity. Sensitivity is improved (to 40–60% in most studies) with the microhematocrit centrifugation technique, which is based on microscopic examination of the buffy coat after centrifugation of four to six microhematocrit tubes. The most sensitive method (~90%) is the miniature anion-exchange centrifugation technique, which is based on the visualization of trypanosomes in eluate after the passage of a large volume (500 μL) of blood through an anion-exchange column and subsequent centrifugation.
As long as treatment of first- and second-stage HAT differs, staging remains an obligatory diagnostic step and is based on the examination of CSF obtained by lumbar puncture. Second-stage HAT is defined by the presence in CSF of a raised leukocyte count (>5/μL) and/or of trypanosomes. The latter can be detected in the cell-counting chamber or, preferably, after centrifugation of the CSF.
Several molecular methods based on PCR or loop-mediated isothermal amplification have been developed, mostly based on the detection of multiple-copy DNA targets of the Trypanozoon group (to which T. brucei belongs) or the single-copy TgsGP gene of T. b. gambiense. None of these methods have been fully validated for diagnostic purposes, and a positive result of their application to blood should be interpreted as suspected rather than confirmed HAT. Molecular methods applied to CSF (to detect biomarkers) have not proven more accurate than classical methods for staging and have yielded false-positive results in a substantial proportion of cases.
The diagnosis of T. b. rhodesiense HAT is usually simpler because parasites are more numerous in body fluids. They can occasionally be visualized in a chancre aspirate. In light of the lack of available serologic tests and the high sensitivity of parasite detection methods in blood, wet mounts, and thin/thick smears (Fig. 222-4), the microhematocrit or other concentration techniques are used for both screening and confirmation. For staging, the definition and methods used are the same as for T. b. gambiense HAT.
Trypanosoma brucei rhodesiense in blood (thin smear, Giemsa stain). (Credit to the DPDx team, U.S. Centers for Disease Control and Prevention, Atlanta.)
TREATMENT Human African Trypanosomiasis
The management of HAT is based on general supportive therapy (e.g., rehydration, pain management), treatment of concomitant infections (e.g., malaria, pneumonia), and antiparasitic treatment. The modalities of antitrypanosomal treatment depend on the Trypanosoma species, the stage of illness, and the presence of contraindications (Table 222-4). T. B. GAMBIENSE
Pentamidine isethionate is highly effective (>95%) against first-stage T. b. gambiense HAT. It is generally well tolerated and can therefore be administered in peripheral health care centers in endemic countries (Fig. 222-5). Hypotension after injection is common but generally mild. Hypoglycemia or hyperglycemia occasionally occurs, but permanent diabetes is very rare. Pain at the injection site is common after intramuscular (IM) injections, but local sterile or bacterial abscesses are rare if basic aseptic precautions are taken. Severe adverse events, such as acute pancreatitis and anaphylaxis, occur extremely rarely.
Nifurtimox–eflornithine combination therapy is also extremely effective (>95% cure rate) and is safer than 14 days of eflornithine monotherapy for treatment of second-stage T. b. gambiense HAT. Common adverse reactions include gastrointestinal disturbances (nausea, vomiting, abdominal pain), headaches, anorexia, and reversible bone-marrow toxicity (anemia, leukopenia). Convulsions and psychosis are reported in fewer than 5% of patients. T. B. RHODESIENSE
Suramin has been used for >90 years and remains the first-line treatment for first-stage T. b. rhodesiense HAT. Common adverse events are pyrexia and nephrotoxicity, which is usually mild and reversible but necessitates surveillance of albuminuria and renal function before each dose.
As eflornithine is ineffective against T. b. rhodesiense, melarsoprol, an arsenic-based derivative, remains the only existing treatment for second-stage T. b. rhodesiense HAT. Reactive encephalopathy is a life-threatening adverse event that occurs in 5–18% of patients, with an associated mortality rate of 10–70%. The efficacy of concomitant high-dose prednisolone to prevent reactive encephalopathy in patients with T. b. rhodesiense HAT is not known. Other severe but less frequent adverse reactions to melarsoprol include exfoliative dermatitis, bloody diarrhea, peripheral neuropathy, renal dysfunction, and liver toxicity. Phlebitis is common, as is soft tissue necrosis if the drug is accidentally given paravenously.
TABLE 222-4Treatment of Human African Trypanosomiasis (HAT) ||Download (.pdf) TABLE 222-4 Treatment of Human African Trypanosomiasis (HAT)
|Disease and Stage ||First-Line Treatment ||Alternative Treatment |
|Drug(S) and Route ||Dose and Duration |
|T. b. gambiense HATa |
|First stage ||Pentamidine isethionate IM or IVa ||4 mg/kg per day for 7 days ||Suramin IVb |
|Second stage ||Eflornithine IV + nifurtimox PO || |
Eflornithine: 200 mg/kg bid for 7 days
Nifurtimox: 5 mg/kg tid for 10 days
|Eflornithine IV: 100 mg/kg qid for 14 daysc |
|T. b. rhodesiense HAT |
|First stage ||Suramin IV ||4–5 mg/kg on day 1 followed by 5 weekly injections of 20 mg/kg (e.g., days 3, 10, 17, 24, 31)d ||Pentamidine isethionate IM or IVe |
|Second stage ||Melarsoprol IV ||2.2 mg/kg per day for 10 days ||— |
Intramuscular injection of pentamidine by a nurse in a village health center, Province Orientale, Democratic Republic of the Congo.
More than 95% of patients with first-stage and second-stage T. b. gambiense HAT are definitively cured with pentamidine and nifurtimox–eflornithine combination therapy, respectively. The overall case–fatality rate is <1% except in very advanced cases. As relapses can occur long after completion of treatment, follow-up visits are advised every 6 months for at least 2 years. The relapse rate is very low (<2%) with current first-line therapies; thus blood and CSF examinations during follow-up visits are no longer standard but can be restricted to symptomatic patients. Patients with second-stage T. b. rhodesiense HAT are at a 5–10% risk of dying during or after melarsoprol treatment, but relapses are very rare.
The elimination of sleeping sickness as a public health problem is in sight, thanks to increased control activities run by national control programs and nongovernmental medical organizations, improved funding, and the end of several civil wars (e.g., in Angola) in the last 15 years. Funding for research, development, and implementation of improved diagnostic methods (e.g., rapid diagnostic tests) and therapeutic tools (e.g., oral drugs) remains crucial to sustain recent achievements.