Advertisement
Research Article| Volume 118, P44-51, May 2022

Leprosy Relapse: A Retrospective Study on Epidemiologic, Clinical, and Therapeutic Aspects at a Brazilian Referral Center

  • Ana Cláudia Mendes do Nascimento
    Affiliations
    Postgraduate Program in Health Sciences, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil
    Search for articles by this author
  • Diogo Fernandes dos Santos
    Affiliations
    National Reference Center for Sanitary Dermatology and Leprosy, Clinics’ Hospital, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil, Postgraduate Program in Health Sciences, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil
    Search for articles by this author
  • Douglas Eulálio Antunes
    Affiliations
    National Reference Center for Sanitary Dermatology and Leprosy, Clinics’ Hospital, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil, Postgraduate Program in Health Sciences, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil
    Search for articles by this author
  • Maria Aparecida Gonçalves
    Affiliations
    National Reference Center for Sanitary Dermatology and Leprosy, Clinics’ Hospital, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil
    Search for articles by this author
  • Marcela Araujo de Oliveira Santana
    Affiliations
    National Reference Center for Sanitary Dermatology and Leprosy, Clinics’ Hospital, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil
    Search for articles by this author
  • Bruno de Carvalho Dornelas
    Affiliations
    Postgraduate Program in Health Sciences, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil

    National Reference Center for Sanitary Dermatology and Leprosy, Clinics’ Hospital, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil, Postgraduate Program in Health Sciences, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil
    Search for articles by this author
  • Luiz Ricardo Goulart
    Affiliations
    National Reference Center for Sanitary Dermatology and Leprosy, Clinics’ Hospital, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil, Postgraduate Program in Health Sciences, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil, Institute of Genetics and Biochemistry, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil
    Search for articles by this author
  • Isabela Maria Bernardes Goulart
    Correspondence
    Corresponding author
    Affiliations
    National Reference Center for Sanitary Dermatology and Leprosy, Clinics’ Hospital, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil, Postgraduate Program in Health Sciences, School of Medicine, Federal University of Uberlândia (UFU), Uberlândia, MG, Brazil
    Search for articles by this author
Open AccessPublished:January 08, 2022DOI:https://doi.org/10.1016/j.ijid.2022.01.009

      Highlights

      • High incidence of cases diagnosed as leprosy relapse.
      • Longer disease-free survival of patients treated with 24 doses of multidrug therapy–multibacillary.
      • Clinical findings and molecular tests should be considered for leprosy relapse.

      Abstract

      Objectives

      We aimed to characterize the profile of patients diagnosed with leprosy relapse and understand the influence of different multidrug therapy (MDT) treatments and initial disease presentation.

      Methods

      This retrospective study included patients diagnosed with leprosy relapse at a referral center in Brazil from 2013 to 2018. We analyzed their clinico-epidemiologic characteristics, laboratory data, and bacilloscopic tests. Survival analysis was used to determine the time elapsed until relapse according to the previous treatment and clinical forms of the disease.

      Results

      A total of 126 cases of relapse were analyzed, which comprised 11.89% (126/1059) of the cases. The median time elapsed until a relapse was 10 years, and most patients had previously undergone 12 doses of MDT (40.48%; 51/126). Undergoing 24 doses of MDT was associated with a better prognosis regarding relapse over time compared with 6 or 12 doses of MDT therapy. Most cases of relapse were classified as multibacillary (96.03%; 121/126).

      Conclusion

      The incidence of relapse was greater than observed in other studies. The high percentage of multibacillary patients who had negative bacillary indices demonstrated that the bacillary index cannot be considered to be an essential criterion for relapse, especially concerning making an early diagnosis.

      Keywords

      Introduction

      Leprosy relapse is defined as patients presenting new clinical signs and symptoms of the active disease after they were adequately treated with a standard therapeutic regimen and discharged as cured (
      • Kaimal S
      • Thappa D.
      Relapse in leprosy.
      ).
      In the 1990s, the World Health Organization (WHO) introduced multidrug therapy (MDT) as a standard treatment for leprosy. The regimen consisted of a combined use of 3 drugs in fixed doses (dapsone, rifampicin, and clofazimine) with varying compositions according to the operational classification (
      World Health Organization
      WHO Expert Committee on Leprosy: Seventh Report.
      ). Patients considered to be paucibacillary (PB) received an MDT regimen of 6 doses (MDT-PB), whereas multibacillary (MB) patients received a regimen of 12-24 doses (MDT-MB) (
      World Health Organization
      WHO Expert Committee on Leprosy: Seventh Report.
      ).
      Previously, after receiving monotherapy of dapsone, MB patients considered cured continued to receive dapsone for 5-10 years, whereas cured PB patients continued for 3-5 years. Nonetheless, the relapse rates were extremely high: a relapse rate of 13.2% among MB patients 6 years after the end of treatment was reported (
      • Becx-Bleumink M.
      Relapses in leprosy patients after release from dapsone monotherapy; experience in the leprosy control program of the all Africa Leprosy and Rehabilitation Training Center (ALERT) in Ethiopia.
      ). In treatments using MDT in fixed-dose regimens, patients are considered “cured” after completing the number of doses appropriate for their operational classification (
      • Girdhar BK
      • Girdhar A
      • Kumar A.
      Relapses in multibacillary leprosy patients: effect of length of therapy.
      ). This approach is unique among treatments for infectious diseases because patients are considered as cured independent of whether signs and symptoms persist and whether elimination of the bacillus has been proven (
      • Kaimal S
      • Thappa D.
      Relapse in leprosy.
      ).
      After MDT was established, both the prevalence of the disease and relapse rate diminished considerably (
      World Health Organization
      WHO Expert Committee on Leprosy: Seventh Report.
      ). According to WHO, the relapse rate among patients treated using standard MDT is low (
      World Health Organization
      ). However, there has been much debate about the real impact of this reduction, given that the numbers of new cases continue to be high in endemic countries (
      • Guerrero-Guerrero MI
      • Muvdi-Arenas S
      • León-Franco CI.
      Relapses in multibacillary leprosy patients: a retrospective cohort of 11 years in Colombia.
      ;
      • Balagon MF
      • Cellona RV
      • Ed Cruz
      • Burgos JA
      • Abalos RM
      • Walsh GP
      • et al.
      Long-Term Relapse Risk of Multibacillary Leprosy after Completion of 2 Years of Multiple Drug Therapy (WHO-MDT) in Cebu, Philippines.
      ) and that many studies have found high relapse rates (
      • Girdhar BK
      • Girdhar A
      • Kumar A.
      Relapses in multibacillary leprosy patients: effect of length of therapy.
      ;
      • Gelber RH
      • Balagon VF
      • Cellona RV.
      The relapse rate in MB leprosy patients treated with 2-years of WHO-MDT is not low.
      ;
      • Cellona RV
      • Balagon MFV
      • dela Cruz EC
      • Burgos JA
      • Abalos RM
      • Walsh GP
      • et al.
      Long-term efficacy of 2 year WHO multiple drug therapy (MDT) in multibacillary (MB) leprosy patients.
      ). These data place doubt on the real effectiveness of the disease control program and MDT regimen.
      To diagnose a leprosy relapse, the possibility of therapeutic failure or insufficiency needs to be ruled out. The lack of uniformity of these concepts, the use of different indicators for assessing relapses (
      • Girdhar BK
      • Girdhar A
      • Kumar A.
      Relapses in multibacillary leprosy patients: effect of length of therapy.
      ;
      • Kaimal S
      • Thappa D.
      Relapse in leprosy.
      ;
      • Visschedijk J
      • Broek J
      • Eggens H
      • Lever P
      • Beers S
      • Klatser P.
      Review: Mycobacterium leprae - millennium resistant! Leprosy control on the threshold of a new era.
      ;
      • Diniz LM
      • Moreira MV
      • Puppin MA
      • Oliveira MLWDR de
      Estudo retrospectivo de recidiva da hanseníase no Estado do Espírito Santo.
      ;
      • Guerrero-Guerrero MI
      • Muvdi-Arenas S
      • León-Franco CI.
      Relapses in multibacillary leprosy patients: a retrospective cohort of 11 years in Colombia.
      ;
      • Balagon MF
      • Cellona RV
      • Ed Cruz
      • Burgos JA
      • Abalos RM
      • Walsh GP
      • et al.
      Long-Term Relapse Risk of Multibacillary Leprosy after Completion of 2 Years of Multiple Drug Therapy (WHO-MDT) in Cebu, Philippines.
      ), and the absence of up-to-date clinical and laboratory criteria for diagnosing relapse makes it difficult to compare different studies and impedes standardization of control programs. In the light of WHO's worldwide strategy for controlling leprosy and the adoption of relapses as an indicator for the effectiveness of the program (

      World Health Organization. Global Leprosy Strategy 2016–2020. Accelerating towards a leprosy-free world. Monitoring and Evaluation Guide. 2017.

      ), the current study aimed to characterize patients who had been diagnosed as presenting relapse of leprosy at a national referral center in Brazil and determine the influence of the initial clinical form of the disease and its treatment on the time that elapsed until relapse.

      Methods

      Study design and participants

      This was a retrospective study conducted in a national referral center for leprosy in Brazil. Patients with a diagnosis of leprosy relapse who were being treated from 2013 to 2018 were identified. Erroneous or duplicated notifications were excluded. In total, 126 patients were included: their data were obtained from the database of the Notifiable Diseases Information System (Sistema de Informação de Agravos de Notificação, SINAN) and medical files.
      This study was approved by the ethics committee of the Federal University of Uberlândia as number 60427816.4.0000.5152. Each participant signaled their acceptance of participation by signing a free and informed consent statement.

      Definition of relapse

      The criteria used to make the diagnosis of relapse are listed in Table 1. We considered patients with neural relapse to be those with clinical evidence of evolving peripheral neuropathy, with the absence of new skin lesions and negative bacilloscopy results. To exclude the possibility of reinfection from the family, the diagnosis of relapse was considered a sentinel event triggering a search for family contacts. To reduce the risk of an autoimmune phenomenon as a confounding factor, all patients showing signs of leprosy reaction were previously treated with corticoid therapy. When the reaction was refractory to this treatment and the patients presented with clinical and laboratory aspects indicating persistence and proliferation of the bacilli, we diagnosed a relapse and started a new treatment with MDT. We used standard criteria (Table 1) for the relapse diagnosis considering clinical and laboratory data to increase the specificity of the diagnosis.
      Table 1– Definition of leprosy relapse according to the clinical and laboratory criteria
      Definition of relapse
      Paucibacillary (PB) patients whose drug treatment ended 3 or more years previously and multibacillary (MB) patients whose drug treatment ended 5 or more years previously, who present:
      Clinical criteriaAppearance of new skin lesions or activation of preexisting lesions; New areas with altered sensitivity; New neurological alterations, evolving lesions, or reactional outbreaks that do not respond to clinical treatment.¹
      Laboratory criteriaPresence of complete bacilli in slit skin smears or skin biopsies; Maintained or increased bacillary DNA load in qPCR tests; Increased/maintained anti-PGL-I ELISA.
      1 - Treatment with thalidomide or corticosteroid at the doses and duration recommended according to the clinical condition.
      qPCR: real-time polymerase chain reaction.

      Study variables

      Epidemiologic variables. Sex, age, diagnosis (clinical, laboratory, or clinical + laboratory), the time elapsed until relapse (in months), previous treatment.
      Clinical variables. Operational classification: PB/MB at diagnosis of relapse and previous treatment; clinical form according to Ridley-Jopling classification (
      • Ridley DS
      • Jopling WH.
      Classification of leprosy according to immunity. A five-group system.
      ): tuberculoid (TT), borderline tuberculoid (BT), borderline (BB), borderline lepromatous (BL), or lepromatous (LL) at diagnosis of relapse and previous treatment and primary neural leprosy (PNL) at the previous diagnosis and neural relapse at diagnosis of relapse. Incapacity level at relapse: grade 0 (G0): no neural impairment in the eyes, hands, or feet; grade 1 (G1): reduction or loss of sensitivity; grade 2 (G2): the presence of incapacity and deformity such as lagophthalmos, claws, bone reabsorption or hands and feet falling off (
      • Brandsma JW
      • Van Brakel WH.
      WHO disability grading: operational definitions.
      ).
      Laboratory variables. Slit skin smear (real-time polymerase chain reaction, quantitative polymerase chain reaction (qPCR); and bacilloscopic index, BI); skin biopsy (qPCR and BI); anti–phenolic glycolipid I (anti-PGL-I) IgM enzyme-linked immunosorbent assay (ELISA).

      Laboratory analysis

      Skin smear. Skin smears were collected from at least 6 points: each earlobe, elbow, and knee and any skin lesions, if present.
      Histopathological skin examination. This procedure began with antisepsis of the lesion using iodized alcohol, followed by subcutaneous local anesthesia using 2% lidocaine. A wedge-shaped incision was made using a scalpel blade, and a fragment of approximately 1 cm along its greatest length that reached the hypodermis was removed. One part of the skin sample to be sent to the molecular pathology and biotechnology laboratory was wrapped in sterilized aluminum paper and immersed in liquid nitrogen. The other part was sent to the institution's pathology laboratory in a flask containing 10% buffered formalin, for histopathological evaluation. Fite-Faraco staining was used to investigate Mycobacterium leprae.
      Bacilloscopy. The bacilloscopic analysis was performed on the skin smear samples described. The BI was determined by the number of bacilli viewed per field in sections made from the samples and was graded from 0 to 6 (
      • Ridley DS
      • Jopling WH.
      A classification of leprosy for research purposes.
      ).
      Anti-PGL-I ELISA serological test. The detection of IgM serum antibodies through ELISA was performed against purified native PGL-I obtained from the cell wall of M. leprae (
      • Lobato J
      • Costa MP
      • Reis EDM
      • Gonçalves MA
      • Spencer JS
      • Brennan PJ
      • et al.
      Comparison of three immunological tests for leprosy diagnosis and detection of subclinical infection.
      ). The ELISA index was considered positive when ≥1.0.
      DNA extraction and real-time qPCR. DNA extraction was performed on the skin smear and skin biopsy samples. To detect the DNA of M. leprae, a primer for qPCR was used. The aim was to ascertain the dispersed-repetition species-specific genomic element (RLEP) of M. leprae. The ABI 7300 real-time PCR system was used (Applied Biosystems, Foster City, CA, USA). The flasks were processed and compared with 2 negative controls to rule out contamination.

      Statistical analysis

      The nonparametric Kruskal-Wallis test was used to compare differences in the mean rank sum among the laboratory variables, in relation to the different clinical forms.
      The time that elapsed until relapse (from the end of the first MDT treatment until the outcome of relapse) was the main dependent variable. Thus, the Kaplan-Meier method was used to compare the differences between the survival curves of the different clinical forms. The first treatment and its influences on the main outcome were examined using the log-rank, Breslow, and Tarone-Ware tests. The SPSS statistical software, version 22, was used. The significance level was taken to be 5% (0.05) in all the analyses. To evaluate the previous treatment, only the patients who had undergone standard MDT were considered.

      Results

      Over the period studied, 126 cases of relapse were confirmed. This corresponded to 11.89% of leprosy cases notified in our clinic over the same period (126/1059). All 126 patients were tested for drug resistance because it is a possible mechanism for leprosy relapse. We identified 11 cases of resistance. Of those, 8 patients presented rifampin resistance, and 3 patients presented rifampin and dapsone resistance. The absolute number of cases of relapse increased progressively, from 8 patients in 2013 to 31 in 2018. During this period, 82 cases of therapeutic insufficiency (7.74%; 82/1059) and 50 cases of therapeutic failure (4.72%; 50/1059) were diagnosed.
      Information relating to the first diagnosis and treatment that was available in the medical files is listed in Table 2. The majority of cases were classified as multibacillary (83.3%; 105/126), and the predominant clinical form was LL (32.5%; 41/126), followed by BL (13.5%; 17/126).
      Table 2Clinical characteristics of the initial treatment in patients who were diagnosed as presenting leprosy relapse during 2013-2018
      Variables%
      Initial clinical form
      PNL32.4
      TT64.8
      BT32.4
      BB86.3
      BL1713.5
      LL4132.5
      NA4838.1
      Operational classification
      PB1310.3
      MB10583.3
      NA86.3
      Previous treatment
      MDT 6 doses1411.1
      MDT 12 doses5140.5
      MDT 24 doses4434.9
      Monotherapy 12 months21.6
      Monotherapy 24 months32.4
      Others/not identified129.5
      TT, tuberculoid; PNL, primary neural leprosy; BT, borderline tuberculoid; BB, borderline; BL, borderline lepromatous; LL, lepromatous; PB, paucibacillary; MB, multibacillary; NA, not possible to assess; MDT, multidrug therapy.
      For the majority of patients (86.5%; 109/126), initial therapy consisted of standard MDT. The regimen most used was 12 doses of MDT/MB (40.5%; 51/126). In the cases treated with MDT/MB in 12 doses, 2 were PNL, 1 BT, 3 BB, 14 BL, 13 LL, and 18 did not have this information available. In the patients who underwent MDT in 6 doses (11.11%; 14/126), 6 were TT, 1 PNL, 2 BT, 1 BB and 4 did not have any data available. Among those who underwent MDT/MB in 24 doses, 25 were LL, 3 were BL, 4 were BB, and 12 did not have information available. Monotherapy using dapsone was administered to 4% of the patients (5/126), in whom the time that had elapsed until a relapse was >20 years. Only 1 monotherapy case had an identifiable clinical form, which was LL. The remaining patients (9.5%; 12/126) underwent other regimens or did not have information available.
      The patients’ characteristics at diagnosis of relapse are listed in Table 3. We observed a slight predominance of the male sex (54%; 68/126) and the patients’ mean age at diagnosis was 49.5 years (± 14.25). Two patients aged <15 years were diagnosed with a relapse.
      Table 3Clinical and epidemiologic characteristics of patients diagnosed with leprosy relapse between 2013 and 2018
      VariablesN%
      Sex
      Male6854
      Female5846
      Age at the time of relapse (years)
      Mean ± SD49.5 (± 14.25)
      Clinical form of relapse
      Neural86.3
      BT5140.5
      BB97.1
      BL118.7
      LL4737.3
      Operational classification of relapse
      PB54
      MB12196
      Degree of incapacity at relapse
      G05543.7
      G13326.2
      G23225.4
      Time until relapse (months)
      Median120
      Diagnosis
      Clinical4233.33
      Laboratory2822.22
      Clinical + laboratory5644.44
      BT, borderline tuberculoid; BB, borderline; BL, borderline lepromatous; LL, lepromatous; PB, paucibacillary; MB, multibacillary; G0, grade zero incapacity; G1, grade 1 incapacity; G2, grade 2 incapacity.
      When cases were categorized by their clinical form, we observed that the BT (40.5%; 51/126) and LL (37.3%; 47/126) forms predominated. Most of the cases (96%; 121/126)) were considered to be multibacillary at the time of the relapse. Most of the patients (51.6%; 65/126) already presented some degree of incapacity at this time, with 26.2% (33/126) in grade 1 and 25.4% (32/126) in grade 2. The median length of time between the end of the first treatment and relapse was 120 months (range 36-432 months) (Table 3).
      The tests performed to make the diagnosis of relapse are listed in Table 4. The anti-PGL-I IgM ELISA test was positive in most of the cases of LL (91.3%; 42/46), such that the highest means were found in relation to the BL form (3.29 ± 2.64) and the LL form (3.24 ± 2.23) (p < 0.001). Among patients with leprosy who presented negative ELISA tests at relapse, 2 were using oral corticoids that may have influenced the result.
      Table 4Laboratory characteristics of patients diagnosed with leprosy relapse according to the clinical form
      Laboratory testsClinical form at relapse
      NeuralBTBBBLLL
      (n = 8)(n = 51)(n = 9)(n = 11)(n = 47)
      Anti-PGL-I ELISAn = 8n = 51n = 9n = 11n = 46
      Positivityb no. (%)2 (25)28 (54.9)7 (77.8)8 (72.7)42 (91.3)
      ELISA index, mean (SD)1.16 (±1.58)1.59 (± 1.37)2.09 (±0.99)3.29 (±2.64)3.24 (±2.23)<0.001a
      qPCR skin biopsyn = 8n = 47n = 9n = 11n = 45
      Positivity no. (%)2(25)25 (53.19)6 (66.7)10 (90.9)39 (86.7)
      Quantitative DNA - mean (SD)120 (±3.17 × 10²)1.21 × 1011 (±7.87 × 10¹¹)7.33 × 106 (±2.05 × 107)5.63 × 107 (±1.67 × 108)7.96 × 107 (±2.94 × 108)<0.001a
      qPCR slit skin smearn = 8n = 45n = 8n = 9n = 44
      Positivity no. (%)1 (12.5)28 (62.2)2 (25)7 (77.8)30 (68.1)
      Quantitative DNA - mean (SD)3.14 × 10² (±7.15 × 10²)7.05 × 104 (±3.33 × 105)1.74 × 10³ (±3.25 × 10³)7.03 × 104 (±1.41 × 105)1.98 × 107 (±8.28 × 107)<0.004a
      BI skin biopsyn = 8n = 50n = 8n = 11n = 45
      Positivity no. (%)07 (14)3 (37.5)6 (54.5)25 (55.55)
      MBI slit skin smearn = 8n = 50n = 9n = 11n = 44
      Positivity no. (%)03 (6)2 (22.22)4 (36.36)19 (43.18)
      mean (SD)00.02 (±0.12)0.07 (±0.13)0.83 (±1.62)1.30 (±1.85)<0.001a
      Abbreviations: SD, standard deviation; BT, borderline tuberculoid; BB, borderline; BL, borderline lepromatous; LL, lepromatous; qPCR, quantitative polymerase chain reaction; BI, bacilloscopic index; MBI, mean bacilloscopic index.
      a. p value for comparison of means (Kruskal-Wallis test) between the groups.
      b. For the anti-PGL-I values, ELISA index ≥1.0 was considered positive.
      The rate of positivity of qPCR for detecting the DNA of M. leprae from skin biopsies was 68.3% (82/120). Among patients with leprosy, the rate of positivity of this test was high: 87.5% (49/56) for the BL and LL forms together.
      Concerning qPCR on smears, 59.6% (68/114) of the samples were positive. The rate was higher in the lepromatous cases (69.8%; 37/53) for the BL and LL forms together.
      Bacilloscopy performed by means of histopathological examination of skin samples presented the highest positivity in LL cases (55.5%; 25/45), with BL cases in second place (54.5%; 6/11). None of the patients with PNL presented positive results observed from the bacilloscopy samples. Among the MB cases, 73.5% (89/121) presented negative bacillary indices at diagnosis. Among these, bacillary DNA was detected via qPCR on skin smear or skin biopsy samples in 74.15% (66/89).
      The highest percentage positivity and highest means from bacilloscopy on smears were found among LL patients (43.18%; 19/44 and 1.30 ± 1.85, respectively), followed by BL patients (36.36%; 4/11 and 0.83 ± 1.62). None of the patients with PNL presented positivity.

      Survival analysis

      In the analysis on the period between the end of the previous treatment and the time of relapse, the median length of time for those treated with the 24-dose MDT/MB regimen was 168 months (14 years), versus 96 months (8 years) for the patients who completed a 6-dose or 12-dose MDT/PB regimen (Table 5).
      Table 5Mean and median length of time (in months) until relapse according to the type of multidrug therapy regimen in the leprosy patients
      Type of treatmentMeanMedianMinimumMaximum
      MDT-PB 6 dose118.2869636240
      MDT-MB 12 dose117.1769636360
      MDT-MB 24 dose171.27316860432
      General139.15612036432
      Abbreviation: MDT: multidrug therapy.
      Figure 1a shows statistically significant differences in predictions for increased survival according to the type and duration of treatment, according to the Kaplan-Meier method, over the entire period observed (p < 0.001). The greatest incidence of cases occurred between the 5th and 10th years after treatment, with around 10 relapses per year. The curves that represent the patients who underwent 6 and 12 doses of MDT show a worse prognosis for relapse, which can be seen through the higher accumulated incidence, whereas the curve for 24 doses of MDT, with the best prognosis, shows lower accumulated incidence.
      Figure 1
      Figure 1Survival analysis (proportion of patients who presented relapse) versus length of time until recurrence. 1a. Survival curve versus type of MDT. 1b. Survival curve (−1) versus type of MDT. 1c. Graph of accumulated risk for the time until relapse versus type of MDT. p Values <0.001 (log-rank [Mantel-Cox], Breslow [generalized Wilcoxon] and Tarone-Ware tests).
      Figure 1b shows that at 120 months after initial treatment (10 years), 64.28% (9/14) of patients treated with 6 doses of MDT/PB presented relapse, as had 70.58% (36/51) of those treated with 12 doses of MDT. Of note, 38.6% (17/44) of the patients treated with 24 doses of MDT/MB presented relapse. At 180 months (15 years), 85.71% (12/14) of MDT/PB patients, 90.19% (46/51) of 12-dose MDT/MB patients and 63.6% (28/44) of the 24-dose patients presented relapse. Patients treated with 24-dose MDT/MB presented a longer interval between the end of the initial treatment and the diagnosis of relapse, as shown by the pink line of the graph, which remained separated to the right throughout the observation period, thus denoting a lower percentage relapse per unit time after the treatment. However, 90% (39/44) of these patients presented relapse within 24 years.
      In the evaluation of the accumulated risk for relapse (Figure 1c), the risk was lower in patients treated with 24 doses and increased with time for all patients, especially patients treated with 12 doses (as shown by the blue line).
      Considering the initial clinical form in patients who presented relapse, most (82%; 64/78) showed relapse within 15 years after initial treatment. The mean length of time until a relapse was shortest in BT patients (60 months), followed by the cases of PNL (104 months). The longest mean times until relapse was observed in the borderline cases (150 months) and lepromatous cases (144 months) (Table 6).
      Table 6Mean and median lengths of time (in months) until relapse according to the initial clinical form of leprosy in patients with relapse of the disease
      Initial clinical formMeanMedianMinimumMaximum
      TT1307260240
      PNL10412048144
      BT60723672
      BB15012060300
      BL114.3539660240
      LL14410860432
      General132.45610836432
      Abbreviations: TT: tuberculoid; PNL: primary neural leprosy; BT: borderline tuberculoid; BB: borderline; BL: borderline lepromatous; LL: lepromatous
      On the survival curve (Figure 2), patients with an initial diagnosis of BT presented the shortest time until relapse. Two were classified as PB and treated with 6 doses of MDT-PB as the initial treatment, and 1 was classified as MB and treated with 12 doses of MDT-MB. At 120 months, 100% of the BT and patients with PNL presented relapse; 50% (4/8) of BB patients; 58.8% (10/17) of BL patients; 61.9% (26/42) of LL patients; and 66.7% (4/6) of TT patients. At 180 months, these proportions had become 82.35% (14/17) of BL patients, 75% (6/8) of BB patients, and 78.6% (33/42) of LL patients. At 240 months, it became 87.5% (7/8) of BB patients, 100% (17/17) of BL patients, 88.09% (37/42) of LL patients, and 100% (6/6) of TT patients.
      Figure 2
      Figure 2Survival curve (proportion of patients who presented recurrence) versus length of time until recurrence. 2a. Accumulated survival curve versus initial clinical form. 2b. Survival curve (−1) versus initial clinical form. 2c. Accumulated risk of the time until recurrence grouped according to the initial clinical form. p Values <0.03 for log-rank (Mantel-Cox) test, p < 0.05 for Breslow (generalized Wilcoxon) test, and p < 0.04 for Tarone-Ware test.
      The accumulated risk of relapse was highest in tuberculoid patients over the first 5-10 years after initial treatment and highest among patients with leprosy after 200 months. In 10-16 years after initial treatment, the BB, BL, and LL clinical forms presented similar risks.

      Discussion

      We observed a high incidence of cases diagnosed as leprosy relapse (11.89%; 126/1059). Through the correlation that we made between the high proportion of MB patients with relapse (96%; 121/126) and the long period until relapse manifested (mean ∼12 years), our understanding was that the transmission chain of the bacillus has been maintained despite reduction efforts.
      WHO, which considers relapse as an indicator of leprosy treatment effectiveness, defines relapse in MB patients as multiplication of the bacillus in patients who had previously been treated, presumed through an increase in the bacillary index of at least 2+ to the previous value at a single site, with or without associated clinical evidence, such as new skin lesions or neural damage (

      World Health Organization. Global Leprosy Strategy 2016–2020. Accelerating towards a leprosy-free world. Monitoring and Evaluation Guide. 2017.

      ).
      Many of our MB patients presented negative bacillary indices at the time of relapse diagnosis. However, the presence of M. leprae DNA at the examined sites had been identified through tests of higher sensitivity (qPCR and anti-PGL-I IgM ELISA), and high bacillary loads were observed through serological tests (
      • Azevedo M de CS
      • Ramuno NM
      • Fachin LRV
      • Tassa M
      • Rosa PS
      • Belone A de FF
      • et al.
      qPCR detection of Mycobacterium leprae in biopsies and slit skin smear of different leprosy clinical forms.
      ;
      • Cho SN
      • Cellona RV
      • Fajardo TT
      • Abalos RM
      • dela Cruz EC
      • Walsh GP
      • et al.
      Detection of phenolic glycolipid-I antigen and antibody in sera from new and relapsed lepromatous patients treated with various drug regimens.
      ). Positive bacillary indices correlated more accurately with the LL form, which is associated with the presence of greater quantities of bacilli in the skin (
      • Premalatha P
      • Renuka I
      • Meghana A
      • Devi S
      • Charyulu P
      • Sampoorna G.
      Utility of bacillary index in slit skin smears in correlation with clinical and histopathological alterations in Hansen's disease: An attempt to revive a simple useful procedure.
      ).
      Early detection makes it possible to diagnose less bacilliferous clinical forms with a lower potential for transmission or development of incapacities (
      • Bazan-Furini R
      • Motta ACF
      • Simão JCL
      • Tarquínio DC
      • Marques Jr, W
      • Barbosa MHN
      • et al.
      Early detection of leprosy by examination of household contacts, determination of serum anti-PGL-1 antibodies and consanguinity.
      ). For the detection strategy outlined by the WHO to become established (WHO, 2004), it is essential to update the concept of leprosy relapse to consider other clinical findings and molecular tests and eliminate positive bacilloscopy results as an essential criterion.
      The lack of standard criteria and indices for evaluating relapse in different health care services makes it difficult to compare the number of cases. Many authors have adopted the WHO's relapse definition (
      • Girdhar BK
      • Girdhar A
      • Kumar A.
      Relapses in multibacillary leprosy patients: effect of length of therapy.
      ;
      • Balagon MF
      • Cellona RV
      • Ed Cruz
      • Burgos JA
      • Abalos RM
      • Walsh GP
      • et al.
      Long-Term Relapse Risk of Multibacillary Leprosy after Completion of 2 Years of Multiple Drug Therapy (WHO-MDT) in Cebu, Philippines.
      ;
      • Becx-Bleumink M.
      Relapses in leprosy patients treated with multidrug therapy: experience in the leprosy control program of the All Africa Leprosy and Rehabilitation Training Center (ALERT) in Ethiopia; practical difficulties with diagnosing relapses; operational procedures and criteria for diagnosing relapses.
      ). Others have adopted their criteria, in which only a positive bacillary index is considered for the diagnosis (
      • Bona SH
      • Silva LOB do V e
      • Costa UA
      • Holanda AO do N
      • Campelo V
      Recidivas de hanseníase em Centros de Referência de Teresina, Piauí, 2001-2008.
      ), or not even this (
      • Guerrero-Guerrero MI
      • Muvdi-Arenas S
      • León-Franco CI.
      Relapses in multibacillary leprosy patients: a retrospective cohort of 11 years in Colombia.
      ;
      • Ferreira SMB
      • Ignotti E
      • Senigalia LM
      • Silva DRX
      • Gamba MA.
      Recidivas de casos de hanseníase no estado de Mato Grosso.
      ;
      • Prabu R
      • Manickam P
      • Mahalingam VN
      • Jayasree P
      • Selvaraj V
      • Mehendale SM.
      Relapse and deformity in 2177 leprosy patients released from treatment with MDT between 2005 and 2010 in South India: A retrospective cohort study.
      ).
      Another divergence is the minimum time after completion of MDT for considering relapse. Many authors have considered any time after the patient has been released from treatment (
      • Guerrero-Guerrero MI
      • Muvdi-Arenas S
      • León-Franco CI.
      Relapses in multibacillary leprosy patients: a retrospective cohort of 11 years in Colombia.
      ;
      • Prabu R
      • Manickam P
      • Mahalingam VN
      • Jayasree P
      • Selvaraj V
      • Mehendale SM.
      Relapse and deformity in 2177 leprosy patients released from treatment with MDT between 2005 and 2010 in South India: A retrospective cohort study.
      ;
      • Sena IVDO
      • Da Silva Machado R
      • Mendes de Brito BA
      • Evangelista de Araújo TM
      • Freitas da Silva GR
      • Leite Rangel Andrade EM
      Relapsed Cases Of Leprosy In A Hyperendemic City In Northeast Brazil.
      ;
      • Stefani MMA
      • Avanzi C
      • Bührer-Sékula S
      • Benjak A
      • Loiseau C
      • Singh P
      • et al.
      Whole genome sequencing distinguishes between relapse and reinfection in recurrent leprosy cases.
      ), and others have imposed a minimum interval of 3 years (
      • Diniz LM
      • Moreira MV
      • Puppin MA
      • Oliveira MLWDR de
      Estudo retrospectivo de recidiva da hanseníase no Estado do Espírito Santo.
      ). Similar to our study, some authors observed longer times until relapse, establishing a minimum interval of 5 years (
      • Balagon MF
      • Cellona RV
      • Ed Cruz
      • Burgos JA
      • Abalos RM
      • Walsh GP
      • et al.
      Long-Term Relapse Risk of Multibacillary Leprosy after Completion of 2 Years of Multiple Drug Therapy (WHO-MDT) in Cebu, Philippines.
      ;
      • Gelber RH
      • Balagon VF
      • Cellona RV.
      The relapse rate in MB leprosy patients treated with 2-years of WHO-MDT is not low.
      ;
      • Kaimal S
      • Thappa D.
      Relapse in leprosy.
      ).
      Because M. leprae has a long incubation period and relapse occurs through the persistence of “hibernating” bacilli, it does not seem to be appropriate to call relapse a manifestation of symptoms seen only a short time after treatment (
      • Kaimal S
      • Thappa D.
      Relapse in leprosy.
      ;
      • Sasaki S
      • Takeshita F
      • Okuda K
      • Ishii N.
      Mycobacterium leprae and leprosy: a compendium.
      ). In such cases, it is more likely that treatment failed or was insufficient (
      • Kaimal S
      • Thappa D.
      Relapse in leprosy.
      ).
      Our results draw attention to the possible underestimation of the number of cases of relapses after MDT. The low rates found in many studies may be related to short follow-up times (
      • Butlin CR
      • Pahan D
      • Maug AKJ
      • Withington S
      • Nicholls P
      • Alam K
      • et al.
      Outcome of 6 months MBMDT in MB patients in Bangladesh- preliminary results.
      ;
      • Prabu R
      • Manickam P
      • Mahalingam VN
      • Jayasree P
      • Selvaraj V
      • Mehendale SM.
      Relapse and deformity in 2177 leprosy patients released from treatment with MDT between 2005 and 2010 in South India: A retrospective cohort study.
      ;
      • Stefani MMA
      • Avanzi C
      • Bührer-Sékula S
      • Benjak A
      • Loiseau C
      • Singh P
      • et al.
      Whole genome sequencing distinguishes between relapse and reinfection in recurrent leprosy cases.
      ), or loss of follow-up in the patients (
      • Becx-Bleumink M.
      Relapses in leprosy patients treated with multidrug therapy: experience in the leprosy control program of the All Africa Leprosy and Rehabilitation Training Center (ALERT) in Ethiopia; practical difficulties with diagnosing relapses; operational procedures and criteria for diagnosing relapses.
      ). Results similar to ours were found in long-term follow-up studies or retrospective cohort studies with large samples (
      • Becx-Bleumink M.
      Relapses in leprosy patients treated with multidrug therapy: experience in the leprosy control program of the All Africa Leprosy and Rehabilitation Training Center (ALERT) in Ethiopia; practical difficulties with diagnosing relapses; operational procedures and criteria for diagnosing relapses.
      ;
      • Girdhar BK
      • Girdhar A
      • Kumar A.
      Relapses in multibacillary leprosy patients: effect of length of therapy.
      ;
      • Shen J
      • Yan L
      • Sun P.
      Clinical features of relapse after multidrug therapy for leprosy in China.
      ).
      There were significant differences in survival curves of the different clinical forms. The shorter time until relapse in the BT cases drew our attention. Of the 3 patients with this clinical form, 2 were classified and treated as PB at the first diagnosis, which may have been an error of operational classification that lay behind the relapse of symptoms. One of our patients who had previously been classified as TT presented relapse 20 years after the treatment, which may signal the possibility of reinfection or another error of classification.
      In our survival analysis of MDT regimens, 24-dose MDT-MB presented the lowest percentage relapse, compared with the 12-dose MB and 6-dose PB regimens. Although the mean times until relapse in the latter 2 regimens were shorter (∼10 years), these times demonstrate that follow-ups after release should be maintained over longer periods. The longer time until relapse among the 24-dose MDT-MB patients may represent greater efficacy, in association with anergia of the host and slow replication of the bacillus. Therefore, all patients diagnosed with relapse are treated with MDT for 24 months and followed up for 10 years in our clinic.
      Given the slow replication of the bacillus and the long time required for the development of new disease (
      • Sasaki S
      • Takeshita F
      • Okuda K
      • Ishii N.
      Mycobacterium leprae and leprosy: a compendium.
      ), we believe that MB cases of relapse occur at a minimum of 5 years after treatment. Before this time, if BB, BL, and LL patients who completed a 24-dose MDT-MB regimen continued to present clinical activity and altered laboratory tests, we would consider this as treatment failure. If patients with these clinical forms completed a 12-dose regimen and continued to present activity, we would consider this as treatment insufficiency and prescribe a new 12-dose cycle. For BT/MB patients who completed 12 doses less than 5 years back and experience signs and symptoms without response to anti-reactional treatment, we would consider this as treatment failure.
      In the light of the results from this study and the need to involve patients in epidemiologic surveillance of leprosy after MDT, we propose an algorithm (Figure 3) for diagnosing and following up with patients after release from treatment, with the aim of achieving early recognition of relapses. In this manner, we will be contributing toward the reduction of incapacities and sequelae, in addition to lower dissemination of the bacillus and case reduction.
      Figures 3
      Figures 33.1, 3.2, 3.3. Protocol for monitoring patients with leprosy and definitions for relapse, therapeutic failure, and treatment insufficiency. 1- Detailed dermato-neurological clinical examination and motor-sensory evaluation of hands, feet, and eyes. Specific laboratory tests: Anti-PGL-I IgM ELISA persistently positive or increasing; qPCR for M. leprae DNA: persistence or increase of bacillary load in smears or skin or nerve biopsies (when available); Bacilloscopy: maintenance or increase of bacillary index and/or complete bacilli in slit skin smears or skin or nerve biopsies; Electroneuromyography: active neural lesion. 2- Detailed dermato-neurological clinical examination and motor-sensory evaluation of hands, feet, and eyes. Examinations: bacilloscopy on slit skin smears (annual for BB, BL, and LL; for BT, take the clinical condition into consideration), anti-PGL-I IgM ELISA and qPCR for M. leprae DNA, when available.

      Declaration of interests

      None of the authors report any competing interests.

      Funding

      This study was funded by the National Council for Scientific and Technological Development (CNPq), the State Funding Agency of Minas Gerais, Brazil (FAPEMIG), and the Brazilian National Health Fund. The funders had no role in the writing of the manuscript.

      Acknowledgments

      The authors wish to thank Matthew Martin Pavelka for editorial assistance related to the English translation during the preparation of the manuscript.

      References

        • Azevedo M de CS
        • Ramuno NM
        • Fachin LRV
        • Tassa M
        • Rosa PS
        • Belone A de FF
        • et al.
        qPCR detection of Mycobacterium leprae in biopsies and slit skin smear of different leprosy clinical forms.
        Braz J Infect Dis. 2017; 21: 71-78https://doi.org/10.1016/j.bjid.2016.09.017
        • Balagon MF
        • Cellona RV
        • Ed Cruz
        • Burgos JA
        • Abalos RM
        • Walsh GP
        • et al.
        Long-Term Relapse Risk of Multibacillary Leprosy after Completion of 2 Years of Multiple Drug Therapy (WHO-MDT) in Cebu, Philippines.
        Am J Trop Med Hyg. 2009; 81: 895-899https://doi.org/10.4269/ajtmh.2009.09-0189
        • Bazan-Furini R
        • Motta ACF
        • Simão JCL
        • Tarquínio DC
        • Marques Jr, W
        • Barbosa MHN
        • et al.
        Early detection of leprosy by examination of household contacts, determination of serum anti-PGL-1 antibodies and consanguinity.
        Mem Inst Oswaldo Cruz. 2011; 106: 536-540https://doi.org/10.1590/S0074-02762011000500003
        • Becx-Bleumink M.
        Relapses in leprosy patients treated with multidrug therapy: experience in the leprosy control program of the All Africa Leprosy and Rehabilitation Training Center (ALERT) in Ethiopia; practical difficulties with diagnosing relapses; operational procedures and criteria for diagnosing relapses.
        Int J Lepr Other Mycobact Dis. 1992; 60: 421-435
        • Becx-Bleumink M.
        Relapses in leprosy patients after release from dapsone monotherapy; experience in the leprosy control program of the all Africa Leprosy and Rehabilitation Training Center (ALERT) in Ethiopia.
        Int J Lepr Other Mycobact Dis. 1992; 60: 161-172
        • Bona SH
        • Silva LOB do V e
        • Costa UA
        • Holanda AO do N
        • Campelo V
        Recidivas de hanseníase em Centros de Referência de Teresina, Piauí, 2001-2008.
        Epidemiol Serv Saúde. 2015; 24: 731-738https://doi.org/10.5123/S1679-49742015000400015
        • Brandsma JW
        • Van Brakel WH.
        WHO disability grading: operational definitions.
        Lepr Rev. 2003; 74: 366-373
        • Butlin CR
        • Pahan D
        • Maug AKJ
        • Withington S
        • Nicholls P
        • Alam K
        • et al.
        Outcome of 6 months MBMDT in MB patients in Bangladesh- preliminary results.
        Lepr Rev. 2016; 87: 171-182
        • Cellona RV
        • Balagon MFV
        • dela Cruz EC
        • Burgos JA
        • Abalos RM
        • Walsh GP
        • et al.
        Long-term efficacy of 2 year WHO multiple drug therapy (MDT) in multibacillary (MB) leprosy patients.
        Int J Lepr Other Mycobact Dis. 2003; 2003 (071≤0308:LEOYWM≥2.0.CO;2): 308-319https://doi.org/10.1489/1544-581X
        • Cho SN
        • Cellona RV
        • Fajardo TT
        • Abalos RM
        • dela Cruz EC
        • Walsh GP
        • et al.
        Detection of phenolic glycolipid-I antigen and antibody in sera from new and relapsed lepromatous patients treated with various drug regimens.
        Int J Lepr Other Mycobact Dis. 1991; 59: 25-31
        • Diniz LM
        • Moreira MV
        • Puppin MA
        • Oliveira MLWDR de
        Estudo retrospectivo de recidiva da hanseníase no Estado do Espírito Santo.
        Rev Soc Bras Med Trop. 2009; 42: 420-424https://doi.org/10.1590/S0037-86822009000400012
        • Ferreira SMB
        • Ignotti E
        • Senigalia LM
        • Silva DRX
        • Gamba MA.
        Recidivas de casos de hanseníase no estado de Mato Grosso.
        Rev Saúde Pública. 2010; 44: 650-657https://doi.org/10.1590/S0034-89102010000400008
        • Gelber RH
        • Balagon VF
        • Cellona RV.
        The relapse rate in MB leprosy patients treated with 2-years of WHO-MDT is not low.
        Int J Lepr Other Mycobact Dis. 2004; 2004 (72≤493:TRRIML≥2.0.CO;2): 493-500https://doi.org/10.1489/1544-581X
        • Girdhar BK
        • Girdhar A
        • Kumar A.
        Relapses in multibacillary leprosy patients: effect of length of therapy.
        Lepr Rev. 2000; 71https://doi.org/10.5935/0305-7518.20000017
        • Guerrero-Guerrero MI
        • Muvdi-Arenas S
        • León-Franco CI.
        Relapses in multibacillary leprosy patients: a retrospective cohort of 11 years in Colombia.
        Lepr Rev. 2012; 83: 247-260
        • Kaimal S
        • Thappa D.
        Relapse in leprosy.
        Indian J Dermatol Venereol Leprol. 2009; 75: 126https://doi.org/10.4103/0378-6323.48656
        • Lobato J
        • Costa MP
        • Reis EDM
        • Gonçalves MA
        • Spencer JS
        • Brennan PJ
        • et al.
        Comparison of three immunological tests for leprosy diagnosis and detection of subclinical infection.
        Lepr Rev. 2011; 82: 389-401
        • Prabu R
        • Manickam P
        • Mahalingam VN
        • Jayasree P
        • Selvaraj V
        • Mehendale SM.
        Relapse and deformity in 2177 leprosy patients released from treatment with MDT between 2005 and 2010 in South India: A retrospective cohort study.
        Lepr Rev. 2015; 86: 345-355
        • Premalatha P
        • Renuka I
        • Meghana A
        • Devi S
        • Charyulu P
        • Sampoorna G.
        Utility of bacillary index in slit skin smears in correlation with clinical and histopathological alterations in Hansen's disease: An attempt to revive a simple useful procedure.
        Ann Med Health Sci Res. 2016; 6: 181https://doi.org/10.4103/2141-9248.183936
        • Ridley DS
        • Jopling WH.
        Classification of leprosy according to immunity. A five-group system.
        Int J Lepr Other Mycobact Dis. 1966; 34: 255-273
        • Ridley DS
        • Jopling WH.
        A classification of leprosy for research purposes.
        Lepr Rev. 1962; 33: 119-128https://doi.org/10.5935/0305-7518.19620014
        • Sasaki S
        • Takeshita F
        • Okuda K
        • Ishii N.
        Mycobacterium leprae and leprosy: a compendium.
        Microbiol Immunol. 2001; 45: 729-736https://doi.org/10.1111/j.1348-0421.2001.tb01308.x
        • Sena IVDO
        • Da Silva Machado R
        • Mendes de Brito BA
        • Evangelista de Araújo TM
        • Freitas da Silva GR
        • Leite Rangel Andrade EM
        Relapsed Cases Of Leprosy In A Hyperendemic City In Northeast Brazil.
        Int Arch Med. 2017; 10https://doi.org/10.3823/2411
        • Shen J
        • Yan L
        • Sun P.
        Clinical features of relapse after multidrug therapy for leprosy in China.
        Lepr Rev. 2015; 86: 165-169
        • Stefani MMA
        • Avanzi C
        • Bührer-Sékula S
        • Benjak A
        • Loiseau C
        • Singh P
        • et al.
        Whole genome sequencing distinguishes between relapse and reinfection in recurrent leprosy cases.
        PLoS Negl Trop Dis. 2017; 11e0005598https://doi.org/10.1371/journal.pntd.0005598
        • Visschedijk J
        • Broek J
        • Eggens H
        • Lever P
        • Beers S
        • Klatser P.
        Review: Mycobacterium leprae - millennium resistant! Leprosy control on the threshold of a new era.
        Trop Med Int Health. 2000; 5: 388-399https://doi.org/10.1046/j.1365-3156.2000.00572.x
      1. World Health Organization. Global Leprosy Strategy 2016–2020. Accelerating towards a leprosy-free world. Monitoring and Evaluation Guide. 2017.

        • World Health Organization
        Report on sixth meeting of the WHO Technical Advisory Group on Elimination of Leprosy. Geneva, 2004
        • World Health Organization
        WHO Expert Committee on Leprosy: Seventh Report.
        WHO, Geneva1998