Figure 1

Figure 1. Flow diagram for studies included in literature review of Hansen disease attributed to Mycobacterium lepromatosis.

Database searches identified 495 references (Figure 1; Appendix). We identified no additional studies through bibliographic screening, but for completeness, we did include 2 case reports published after our database searches (9,11). After de-duplication and screening by title and abstract, we retained 58 studies, all published in peer-reviewed journals, for full text review; we extracted data from 33 (21 case reports, 12 specimen surveys). We excluded 1 because it was a retrospective review of 9 cases (12) that reported only 1 M. lepromatosis case that was also described in another source (13). Similarly, we excluded a review of cases among refugees and migrants in Italy during 2009–2018 (14) that reported PCR testing of 24 cases, 16 positive for M. leprae and 1 for M. lepromatosis, because the M. lepromatosis case was described in more detail in a case report (15).

Among the 21 case report studies, 14 studies described just 1 case, 6 described 2, and 1 described 6, yielding 32 PCR–positive cases: 21 M. lepromatosis–only, 5 M. leprae–only, and 6 dual infections (Table 1; Appendix). Of patients with M. lepromatosis, 10/27 resided in the United States (7 originally from Mexico), 6 in Mexico, 3 in the Dominican Republic, 2 each in Singapore and Myanmar, and 1 each in Indonesia, Paraguay, Cuba, and Canada. One study from Mexico reported 4 family cases, but only 2 were PCR–confirmed to be M. lepromatosis (9). Twenty-two cases occurred in the Americas and 5 in Asia; Mexico was the country of origin or residence for 13/27 case-patients. One source mentioned 2 case-patients from Costa Rica living in the United States but provided no details (23). Median age of case-patients was 41 years (range 21–86 years) and 63.0% (17/27) were male.

We assessed 13/21 studies as good and 8/21 as fair quality (Appendix). Eight studies did not provide detailed PCR methods (11,15,22,24–26,28,29), but 7 of these referred to laboratories (National Hansen’s Disease Programme; US Centers for Disease Control and Prevention; Ecole Polytechnique Fédérale de Lausanne; Japan Leprosy Research Centre) or involved authors with documented experience in M. lepromatosis diagnostic methods (11,22,24–26,28,29).

The case-patient from the study in which M. lepromatosis was first identified (1) was a patient originally from Mexico residing in the United States who had died from DLL with LP. PCR sequencing of the ≈1,500 bp 16S rRNA gene in acid-fast bacilli from frozen liver autopsy specimens showed that the strain, designated FJ924, matched most closely with M. leprae (BLAST analysis [https://blast.ncbi.nlm.nih.gov/Blast.cgi] of 16S rRNA gene, 1,475/1,506 bp, 97.9% identity) and next most closely with M. haemophilum (1,465/1,505 bp, 97.3%). The researchers obtained archived biopsy specimens from a second patient originally from Mexico, also with DLL and LP, who had died 5 years earlier. Gene sequences from that earlier case, including from the 16S rRNA gene, matched 100% with strain FJ924. On the basis of those findings, the researchers proposed a new species, M. lepromatosis, as a second causal agent of DLL, while speculating that it might also cause lepromatous (LL) and borderline lepromatous (BL) forms of HD (1). The researchers also obtained archived specimens from 2 fatal cases of DLL in Singapore (both case-patients died in 1999) with dual M. lepromatosis/M. leprae infection identified using a mix of species-specific and nonspecific primers matched to GenBank sequences (17).

In another study from Mexico, a sample, Mx1-22, taken from an 86-year-old patient with DLL and LP had rrs, rpoB, sigA, and hsp65 gene sequences identical to FJ924 (16). Subsequent studies used a range of species-specific primers and sequencing, 7 targeting 16S rRNA (13,18–21,23,30), 2 hemN (27,31), and 1 the LPMREP repetitive element (9) to confirm M. lepromatosis infection (Table 1). The oldest archived specimen in which M. lepromatosis was identified was from a US-resident patient originally from Mexico, 50 years of age, treated in Carville, Louisiana, USA, who was diagnosed with DLL with LP in 1963 (19). That patient, who initially sought treatment for a soft tissue sarcoma in the right lower leg, developed overt signs of DLL and LP after radiotherapy and amputation of the leg. Histopathologic review identified chronic HD lesions in the skin, vessels, and nerves surrounding the sarcoma, consistent with DLL. The patient survived to 85 years of age.

Figure 2

Figure 2. Types of Hansen disease and reactional states in reports included in literature review of PCR-confirmed cases attributed to M. lepromatosis. BL, borderline lepromatous leprosy; DLL, diffuse lepromatous leprosy;...

Figure 3

Figure 3. Types of Hansen disease by case-patient sex in reports included in literature review of PCR-confirmed cases attributed to M. lepromatosis. BL, borderline lepromatous leprosy; DLL, diffuse lepromatous leprosy;...

Of the 27 M. lepromatosis case-patients, 15 (55.6%) had DLL, 9 of whom also had LP; 10 had LL, 1 had LP, and 2 had BL (Figure 2). Among those cases, type 2 erythema nodosum leprosum (ENL) HD reactions were reported in 3 cases each of DLL and LL and in both BL cases. Male patients comprised 8/15 DLL and 1/2 BL case-patients but a higher proportion (8/10) of LL case-patients (Figure 3). Median time of evolution from initial symptoms to HD diagnosis was 2 years (range 8 months–12 years); 2 patients had been diagnosed with HD 20 and 28 years earlier. Two patients reported direct contact (hunting, handling, or eating) with armadillos in Mexico (23,25); a third came from a village in Mexico where armadillo meat was consumed, but the patient had not eaten it (20). Two patients in the United States had no known risk factors or exposures other than travel, including to Florida (18), worldwide travel including to Asia, the Caribbean, and the Middle East, and 2 trips to the Pacific coast of Mexico that were short but occurred consistent with a 7–8 year incubation period for HD manifestation (13).

Concomitant or differential diagnoses discussed in the case reports included sarcoidosis (initially treated with steroids) (19), syphilis (borderline positive antinuclear antibodies, treated initially with intramuscular penicillin) (23), rheumatoid arthritis (treated with prednisone and methotrexate 2 years earlier) (13), cutaneous vasculitis (treated with azathioprine and prednisone for >5 years) (30), vasculitis related to drug abuse (27), and acute kidney injury (11). All cases were otherwise consistent with the clinical and histopathologic picture of DLL (32): insidious onset with violaceous erythema developing on the face and lower extremities (may or may not be anesthetic); myxoedema-like aspect with smooth, tense, alopecic skin, progressing to madarosis; earlobe infiltration; rhinitis; nasal septal defects; hypohidrosis; xerotic and scaly skin with ichthyosiform appearance on lower limbs; areas of hypoesthesia and hyperesthesia associated with hypopigmented, atrophic plaques; and impaired sensation in the hands and feet becoming more generalized because of progressive nerve involvement. Histologically, dense histiocytic infiltration in skin and nerves was observed, advancing to endothelial proliferation with thickening of vascular walls, leading to occlusion of small arteries, and invasion of internal organs, indicated by hepatomegaly and splenomegaly.

All multibacillary HD case-patients were treated with multidrug therapy, typically with rifampicin, clofazimine, and dapsone (sometimes substituted with minocycline, clarithromycin, moxifloxacin, or oxfloxacin) for 12 or 24 months, plus corticosteroids (mainly prednisone), thalidomide, or both for ENL and LP. Treatment outcomes were favorable for 10/27 patients at time of reporting, although that group included 2 patients who had no regrowth of eyelashes or eyebrows at 3-year follow-up, 1 of whom also experienced ENL at 3 years (9). One patient whose mild neurologic deficits had resolved at 7 years was still taking thalidomide and prednisone because of new, although sparse, ENL lesions (23).

Eight patients were still receiving treatment or had just completed treatment at time of reporting. Most studies did not assess or report grade of disability. Six deaths were reported, of which 4 were attributed to DLL or sepsis secondary to DLL (1,17). One patient who died was a woman, 86 years of age, who improved after 10 days of treatment and was discharged after 2 weeks in stable condition but then died of unknown causes at home 3 months later (16). A man, 72 years of age, died from lung cancer after 5 months; he was described as having leprosy-like illness because, although 16S rRNA sequencing found a 100% match to M. lepromatosis and the patient manifested neurologic and dermatologic symptoms of LL and rhinorrhea, histopathology did not confirm mycobacteria within peripheral nerves (18).

Five case-patients tested positive only for M. leprae: One was a woman, 43 years of age, a United States resident originally from Nepal, diagnosed with midborderline HD with possible type 1 reaction and possible LP because of an erythematous geographic skin plaque which ulcerated, but was not biopsied (28). Another was a woman, 37 years of age, from Colombia, HIV-positive, diagnosed with LL and LP; the LP was potentially triggered as immune reconstitution inflammatory syndrome following initiation of antiretroviral therapy (22). The remaining 3 patients, from the Dominican Republic, 2 with LL and 1 with BL, were from a series of 6 cases that also included 1 M. lepromatosis–only and 2 dual-infection cases (31).

Quality assessment rated 7/12 specimen survey studies as good and 5/12 as fair quality. All but 1 study (42) reported details of PCR methods. Nine studies (34–42) used skin lesion or biopsy or tissue samples, 2 used both skin and lesion biopsy and skin slit smear specimens (6,33), and 1 did not explicitly state the sources of specimens (Table 2; Appendix) (3). Overall, surveys dated 1968–2020 reported 1,098 PCR–positive M. lepromatosis–only, M. leprae–only, or dual-infection findings from 1,428 specimens. M. lepromatosis was identified in 44.9% (133/296, 26 dual infection) of PCR–positive specimens from patients in Mexico or in the United States but originally from Mexico, 3.8% (5/133, 5 dual infection) of patients from Colombia, 12.5% (10/80, 3 dual infection) of patients from Brazil, and 0.9% (2/224) of patients from the Pacific-Asia region; all 157 specimens from China, 50 from Africa (Mali, Uganda), and 77 from Venezuela were positive only for M. leprae. For patients from Mexico, excluding those resident in the United States, M. lepromatosis was detected in 43.9% (116/264) of PCR–positive specimens, including 25 dual infections. For patients resident in the United States from any country of origin, M. lepromatosis was detected in 16.7% (20/120) of PCR–positive specimens, including 1 dual infection.

The distribution of HD types among 116 M. lepromatosis–only and 13 dual-infection patients was tuberculoid in 7 (5.4%); borderline tuberculoid, midborderline, or borderline lepromatous in 20 (15.5%); LL in 73 (56.6%); and DLL in 29 (22.5%). LP was reported in relation to 14/27 specimens from patients in Mexico or in the United States but originally from Mexico. One patient from Mexico with LL who provided a specimen positive for M. lepromatosis had consumed armadillo meat (40).

Our systematic review identified 27 case reports of HD attributed to PCR–confirmed M. lepromatosis infections. In addition, surveys of specimens from current patients and archived material uncovered 153 cases of M. lepromatosis HD. Most of those infections (60% of case reports, 87% of surveyed specimens) occurred in patients resident in Mexico or in the United States but originally from Mexico. Most (70%) of the case reports described patients with DLL, among whom half manifested LP.

Our findings appear to substantiate the hypothesis that M. lepromatosis is the predominant HD pathogen in Mesoamerica and the Caribbean, and particularly in Mexico, and that it has a strong tendency to cause the DLL form of HD and, indirectly, severe LP reaction. However, there are some important caveats. First, DLL and LP were identified in Mexico in the late 19th Century by physicians Lucio Nájera and Ygnacio Alvarado and were further characterized by Fernando Latapí in 1938 (43,44). Discovery of M. lepromatosis in fatal cases of DLL with LP in 2 patients in the United States who were originally from Mexico (1), combined with the high proportion of HD cases in Mexico that were DLL, with or without LP, might have resulted in disproportionate publication of case reports and specimen surveys focused on this form of HD in this region. Laboratory expertise and resources for detecting M. lepromatosis are also more readily available in Mexico and the United States. However, Mexico is not an HD-endemic country, reporting an average of <200 newly detected cases per year during 2005–2021 (45), mostly in the states of Guerrero, Jalisco, Oaxaca, Sinaloa, and Michoacán (46). Although this annual average represents a relatively small number of cases, in the context of HD elimination, it is a matter of public health concern. In addition, M. lepromatosis has a tendency to cause severe forms including DLL, which with its nonnodular manifestation is prone to diagnosis at later stages; therefore, there remains an immense personal impact on persons affected by the disease.

Current HD multidrug therapies appear to be effective treatments, except in the most severe cases in which patients are at risk of secondary infection. However, evidence on the apparent effectiveness of current multidrug therapy regimens in treating HD caused by M. lepromatosis is constrained by the small number of cases described, their clinical complexity and severity, and lack of follow-up data to characterize long-term outcomes, including permanent disabilities.

Our review showed that M. lepromatosis–caused HD occurs in other countries in the Americas and, sporadically, in Asia and the Pacific. Most notably, 1 in 8 specimens from the south of Brazil were identified as M. lepromatosis. Brazil is an HD-endemic country with ≈20,000 newly detected cases per year. Also, the survey data in our review showed that, when type of HD was reported, a higher percentage of cases attributed to M. lepromatosis were LL (57%) than DLL (23%). Even a small fraction of HD cases in Brazil caused by M. lepromatosis would represent a large number of cases. The clear implication is that a national survey of current and newly detected HD cases in Brazil is needed, ideally using the recently validated M. lepromatosis/M. leprae diagnostic assay (6). Parallel studies in neighboring countries where M. lepromatosis and DLL are perhaps more prevalent, would yield sequence data that could be used to investigate the distribution of M. lepromatosis variants and lineages, including drug-resistant strains, to achieve the same level of understanding as for M. leprae (5,47). Whether M. lepromatosis has a pathogenic tendency toward causing DLL and whether certain population groups are more susceptible to developing this form of HD can be investigated by pathogen and host genomic testing across the HD spectrum and in different populations. From a One Health perspective, we know that zoonotic transmission of M. leprae presents a risk to human health (48), and M. lepromatosis has been detected in red squirrels (Sciurus vulgaris) from the British Isles, including Ireland (49). Given that 4 case-patients with HD caused by M. lepromatosis in our review had direct or indirect contact with wild armadillos, a survey of archived specimens or specimens from freshly caught armadillos in Mexico and Brazil is warranted (50).

The narrowly focused scope, sensitivity, and specificity of lepromatosis as a search term and the relatively few references included in our review give us confidence that all relevant studies were identified. Quality of reporting was good in 61% of included studies and fair in the remainder. The tendency of studies to focus on DLL in Mesoamerica, possibly resulting in observational and publication biases for case reports and sampling bias for surveys, were the main sources of bias in our review, although there were several large studies from other regions with null findings for M. lepromatosis. A key quality item considered for this review was adequate description of PCR methods, which most, but not all, studies provided. Specimen surveys more consistently described PCR methods, including targets and primers, than did case reports, some of which covered time periods during which those methods were still being developed. M. lepromatosis does not manifest only as DLL, but most specimen surveys did not provide clinical data for the patients sampled. Even when HD type was stated, misclassification was possible unless HD specialists or reference centers were involved in diagnosis. Although we cannot entirely preclude the possibility of double counting, we identified only a few cases that were reported twice and contacted the authors of 2 studies when geography and time span suggested that possibility to confirm that there was no overlap (33,34).

It is perhaps remarkable that a new species causing an endemic disease of major public health impact has not prompted larger-scale studies to determine its true prevalence. Even if options for patient management are determined by clinical manifestations of HD rather than its etiologic agents, understanding disease attribution and distribution of a highly pathogenic species are clearly important, and the availability of validated PCR methods enables large-scale epidemiologic studies to be conducted.

In conclusion, clinicians need to be aware that Hansen disease of various forms can be caused by either M. leprae or M. lepromatosis. Current multidrug therapy regimens appear to be effective regardless of infecting species.