Background

The 2022 multinational mpox outbreak in nonendemic countries led to its declaration as a health emergency in July 2022, reaching the global number of 90 618 cases and 157 deaths by 27 September 2023, with Brazil being the second most affected country [1,2]. Cisgender MSM were disproportionately burdened by this outbreak; 30–50% of individuals diagnosed with mpox were living with HIV [3–11]. Generally, people with HIV (PWH) and mpox were older and more often presented with concomitant sexually transmitted infections (STIs), anogenital lesions and proctitis [4,8,9,12]. Although initial mpox descriptions reported no substantial differences in hospitalizations or clinical severity according to HIV status, these individuals were mostly virologically stable and without severe, immunosuppression [3,4,7,10,13].

Recent studies, however, suggest that uncontrolled HIV might impact on mpox-related outcomes, especially if CD4+ cell counts are below 200 cells/μl, suggesting an association between advanced HIV disease and severe mpox presentation [5,8,9,14–17]. This is of special concern in Latin America – the region accounted for 55% of PWH with mpox and CD4+ counts lower than 350 cells/μl; access to prevention and experimental mpox treatment were quite limited [15,18]. Overall, mpox-related hospitalization rates during the 2022 multinational outbreak are up to 10%, mostly related to pain control, secondary bacterial infections, urologic or proctologic complications, odynophagia, keratitis and, rarely, encephalitis, pulmonary involvement or myocarditis [4,5,7,15,17,19–22]. Brazil reported up to 50% of mpox cases and at least three deaths among PWH [4,23]. Despite the country's longstanding commitment to universal access to antiretroviral therapy (ART), structural barriers remain, resulting in a high prevalence of HIV late presentation to care and poor ART adherence. These aspects have worsened after the COVID-19 pandemic and contributed to the high mpox burden [24–26].

This study aimed to characterize mpox-related hospitalizations and further explore the impact of HIV status on mpox outcomes among individuals enrolled at the Evandro Chagas National Institute of Infectious Diseases (INI/Fiocruz), a major referral center for mpox in Rio de Janeiro, Brazil.

Methods Study design and participants

This was a single-center, prospective cohort study that enrolled participants older than 18 years diagnosed with mpox from 12 June to 31 December 2022, at INI/Fiocruz. We offered mpox virus quantitative real-time PCR (qPCR) testing to all individuals with suspected infection, which was performed at the Mpox Reference Laboratory at the Oswaldo Cruz Institute/Fiocruz. A positive result on qPCR from any swab specimens led to case confirmation. Follow-up period was 28 days or until resolution of mpox lesions. Individuals requiring hospitalization were followed at the INI-Fiocruz Inpatient Care Unit; if hospitalization occurred elsewhere, data were collected retrospectively. Procedures of INI's mpox cohort have been previously described [4].

Procedures

We prospectively collected sociodemographic, epidemiological, behavior, clinical and laboratory data onto a standardized case report form. Information was gathered on birth date, gender identity and race according to self-report. In Brazil, smallpox vaccination was compulsory until 1975, so for these analyses we considered those born before 1975 as vaccinated. Systemic signs and symptoms included the presence of fever, asthenia, adenomegaly, myalgia, arthralgia and/or headache. For hospitalized participants, invasive support included mechanical ventilatory support, vasoactive drugs, use of nasoenteric tube or indwelling urinary catheters, need for hemodialysis or either arterial or venous central line. We included data on reasons for hospitalization, length of hospitalization and clinical outcomes. Tecovirimat was requested according to the Ministry of Health's protocol for compassionate use, upon availability, if the individual met criteria of a severe clinical course [27].

At the initial assessment, we offered HIV rapid test (3rd generation) performed according to the Brazilian Ministry of Health algorithm [28]; HIV-RNA viral load; rapid Treponema pallidum test for syphilis screening with subsequent confirmation with nontreponemal testing [veneral disease research laboratory (VDRL)]; molecular detection of Chlamydia trachomatis and Neisseria gonorrheae in rectal swabs (Abbott Real Time Platform); hepatitis B surface antigen rapid test; anti-HCV rapid test for hepatitis C. Active syphilis was defined as VDRL titers equal to or higher than 1 : 8.

For PWH, we assessed CD4+ count (cells/μl) and HIV-RNA viral loads records closest to the onset of mpox symptoms, obtained from either clinical charts or the Brazilian Laboratory Control System database (SISCEL). We adapted a cross-sectional cascade of HIV care including the following stages: HIV-diagnosed (previously known HIV infection at mpox diagnosis); linked to HIV care (at least one record of HIV-related care appointment, laboratory examinations or ART prescription after HIV diagnosis and before mpox assessment); retained in HIV care (ART prescription in the last 6 months or at least two records of HIV-RNA viral load or CD4+ cell count in the last year); on ART (at least one dispensation of antiretroviral drugs in the last 6 months, obtained from the Brazilian Antiretroviral Logistics database – SICLOM); and virologically suppressed (HIV-RNA viral load ≤200 copies/ml) [29].

Statistical analysis

We compared study characteristics of confirmed mpox cases according to HIV status and, among PWH, by hospitalization status (PWH hospitalized vs. PWH not hospitalized). We used the chi-squared test or Fisher's exact test for qualitative variables and the Moods median test for quantitative variables. We performed a sub-analysis among PWH to compare clinical outcomes according to CD4+ count strata and to describe HIV continuum of care outcomes by hospitalization status. All analyses were performed in R-Project (4.2.1).

Ethical considerations

This study was approved by the Ethics Review Board at INI-Fiocruz (CAAE #61290422.0.0000.5262). Participants provided written informed consent.

Results

From June 12 to 31 December 2022, 418 participants had a confirmed mpox diagnosis. Overall, median age was 33 years [interquartile range (IQR) 28–40]; 91.6% self-identified as cisgender men, 5.7% as cisgender women and 2.6% as travesti or transgender women (TGW). The majority were black or pardo (60.5%, n = 207/342), 38.6% were white (n = 132/342) and 0.9% were indigenous (n = 3/342). Most participants had studied until primary education (60.2%, n = 213/352), 6% had secondary education (n = 21/352) and 33.8% went to postsecondary education (n = 119/352). Among the cisgender men, 87% were MSM (n = 336/362).

Data on HIV serostatus were available for 409 participants, of whom 52.1% were PWH, mostly with previous HIV diagnosis. Compared with HIV-negative individuals, PWH diagnosed with mpox were older [35 (IQR 30–40) vs. 31 (IQR 26–38), P < 0.01] and most frequently self-reported being cisgender men (95.8 vs. 87.3%, P < 0.01), travesti or TGW (3.3 vs. 2%, P < 0.01) (Table 1). Among those who were negative for HIV, PrEP use was reported by 33% (Table 1).

Table 1 - Sociodemographic, behavioral and clinical characteristics of mpox cases according to HIV status (N = 409). HIV− (N = 196)a [n (%)] HIV+ (N = 213)a [n (%)] Total (N = 409)a [n (%)] P valueb Age (median, IQR) 31 (26–38) 35 (30–40) 33 (28–40) <0.01 Age range <0.01  18–24 years 34/196 (17.4%) 12/213 (5.6%) 46/409 (11.2%)  25–29 years 49/196 (25%) 33/213 (15.5%) 82/409 (20.1%)  30–39 years 71/196 (36.2%) 105/213 (49.3%) 176/409 (43%)  ≥40 years 42/196 (21.4%) 63/213 (29.6%) 105/409 (25.7%) Gender identity <0.01  Cisgender men 171/196 (87.3%) 204/213 (95.8%) 375/409 (91.7%)  Cisgender women 21/196 (10.7%) 2/213 (0.9%) 23/409 (5.6%)  Nonbinary 0/196 0/213 0/409  Transgender men 0/196 0/213 0/409  Travesti or transgender women 4/196 (2%) 7/213 (3.3%) 11/409 (2.7%) Race 0.65  Black 53/166 (31.9%) 52/175 (29.7%) 105/341 (30.8%)  Pardo or mixed 45/166 (27.1%) 57/175 (32.6%) 102/341 (29.9%)  White 67/166 (40.4%) 64/175 (36.6%) 131/341 (38.4%)  Indigenous 1/166 (0.6%) 2/175 (1.1%) 3/341 (0.9%) Education 0.19  Primary 110/169 (65.1%) 100/180 (55.6%) 210/349 (60.2%)  Secondary 9/169 (5.3%) 12/180 (6.6%) 21/349 (6%)  Post secondary 50/169 (29.6%) 68/180 (37.8%) 118/349 (33.8%) MSM 144/165 (87.3%) 190/193 (98.4%) 334/358 (93.3%) <0.01 Current sex work 6/178 (3.4%) 6/190 (3.2%) 12/368 (3.3%) 0.91 PrEP use 64/194 (32.9%) NA NA Reported sex in the last 30 days 175/186 (94.1%) 175/197 (88.8%) 350/383 (91.4%) 0.07 Time from symptoms onset to initial assessment (median, IQR) 6 (4, 9) 6.0 (4, 10) 6.0 (4, 9) 0.52 Median number of sex partners in the last 30 days (IQR) 2 (1, 3) 2 (1, 3) 2 (1, 3) 0.54 Reported anal sex in the last 30 daysc 78/175 (44.6%) 88/175 (50.3%) 166/350 (47.4%) 0.28 Sex contact with potential mpox cased 35/171 (20.5%) 35/173 (20.2%) 70/344 (20.3%) 0.58 Home contact with potential mpox casese 20/169 (11.9%) 12/172 (7%) 32/341 (9.4%) 0.12 Probable vaccinated for smallpox 18/196 (9.2%) 19/213 (8.9%) 37/409 (9%) 0.97 Active syphilis 18/180 (10%) 59/198 (29.8%) 77/383 (20.1%) <0.01 Anorectal gonorrhea 11/141 (7.8%) 15/164 (9.15%) 26/305 (8.52%) 0.69 Anorectal Chlamydia 12/141 (8.5%) 18/164 (11%) 30/305 (9.8%) 0.47 Any bacterial STI 39/180 (21.7%) 84/198 (42.4%) 123/378 (32.5%) <0.01 Hepatitis B 2/181 (1.1%) 3/189 (1.6%) 5/370 (1.4%) >0.99 Hepatitis C 5/187 (2.7%) 19/194 (9.8%) 23/381 (6%) <0.01 Systemic symptoms or signs 157/186 (75.3%) 187/207 (84.1%) 344/393 (79.8%) 0.07 Fever 107/192 (55.7%) 150/211 (71.1%) 257/403 (63.7%) <0.01 Anogenital lesions 143/193 (74.1%) 170/207 (82.1%) 313/400 (78.3%) 0.05 Clinical signs of proctitis 37/194 (19.1%) 63/212 (29.7%) 100/406 (24.6%) 0.01 Rectal swab 0.01  PCR MPXV detectable 80/121 (66.1%) 131/163 (80.4%) 211/284 (75.3%)  PCR MPXV not detectable 41/121 (33.9%) 32/163 (19.6%) 73/284 (24.7%) Required hospitalization 19/196 (9.7%) 24/213 (11.3%) 43/409 (10.5%) 0.60 Death 0/196 2/213 (1%) 2/409 (0.5%) 0.50

IQR, interquartile range; PrEP, pre-exposure prophylaxis.

an (%); median (IQR).

bFisher's exact test; Wilcoxon rank sum test; Pearson's chi-squared test.

cBoth receptive and insertive anal sex.

dReported sexual contact with someone who was suspect case or diagnosis with mpox or presenting mpox-like lesions in the 30 days previous the study entry.

eReported living in the same household of suspected/confirmed mpox case in the 30 days previous the study entry.

Compared with HIV-negative participants, more PWH reported being MSM (98.4 vs. 87.3%, P < 0.01); had active syphilis (29.8 vs. 10%, P < 0.01), hepatitis C (9.8 vs. 2.7%, P < 0.1) or any bacterial STI (42.4 vs. 21.7%, P < 0.01) (Table 1); presented more frequently with fever (71.1 vs. 55.7%, P < 0.01), anogenital lesions (82.1 vs. 74.1%, P = 0.05) and proctitis (29.7 vs. 19.1%, P = 0.01) and more often had a positive MPXV qPCR in rectal swabs (80.37 vs. 66.12%, P = 0.01).

Overall, 10.5% (n = 43) of participants were hospitalized because of mpox (39 at INI-Fiocruz Inpatient Care Unit), 55.8% were PWH, with a median age of 31 years old (IQR 28–39); 93% were cisgender men, 50% self-identified as black/pardo and 53.5% had primary education. The hospitalization rate during follow-up was not different based on HIV status although 34.9% had been admitted at the time of the initial mpox assessment. No participants reported immunosuppression related to drugs or other medical conditions. Median interval between symptoms onset and hospitalization was 9 days (IQR 7–12) and length of hospitalization was 7.5 days (IQR 5–12). The most common reason for hospitalization was pain control (90.7%), often requiring opioids use (67.5%), followed by secondary bacterial infections (55.8%). The majority of hospitalized participants presented with fever (69.1%) and/or anogenital lesions (67.4%). PWH presented more frequently with proctitis [58.33 vs. 26.32%, P = 0.04)] and required invasive support (34.8 vs. 0, P = 0.01) (Table 2).

Table 2 - Sociodemographic, laboratorial, and clinical information on mpox-related hospitalized participants according to HIV status (N = 43). HIV − (N = 19)a HIV + (N = 24)a Total (N = 43)a P valueb Median age (IQR) 30 (25–36) 35 (30–42) 31 (28–40) 0.08 Age range 0.20  18–24 years 3/19 (15.8%) 1/24 (4.2%) 4/43 (9.3%)  25–29 years 6/19 (31.6%) 3/24 (12.5%) 9/43 (20.9%)  30–39 years 7/19 (36.8%) 12/24 (50%) 19/43 (44.2%)  ≥40 years 3/19 (15.8%) 8/24 (33.3%) 11/43 (25.6%) Gender identity 0.08  Cisgender men 16/19 (84.2%) 24/24 (100%) 40/43 (93%)  Cisgender women 3/19 (15.8%) 0/24 3/43 (7%) Race >0.99  White 7/18 (38.9%) 9/21 (42.9%) 16/39 (41.1%)  Pardo or mixed 5/18 (27.8%) 5/21 (23.8%) 10/39 (25.6%)  Black 6/18 (33.3%) 7/21 (33.3%) 13/39 (33.3%) Education 0.13  Primary 13/19 (68.4%) 10/24 (41.7%) 23/43 (53.5%)  Secondary 1/19 (5.3%) 1/24 (4.1%) 2/43 (4.6%)  Postsecondary 5/19 (26.3%) 13/24 (54.2%) 18/43 (41.9%) Fever 14/19 (73.7%) 15/23 (65.2%) 29/42 (69%) 0.55 Anogenital lesions 9/19 (47.4%) 20/24 (83.3%) 29/43 (67.4%) 0.01 Time from symptoms onset to hospitalization (days, IQR) 9 (8–11) 8 (7–14) 9 (7–12) >0.99 Length of hospitalization (days, IQR) 7 (5–10) 8 (4–14) 7.5 (5–12) 0.56 Reason of hospitalization  Pain control 18/19 (94.7%) 21/24 (87.5%) 39/43 (90.7%) 0.62  Urological complication 7/19 (36.8%) 10/24 (41.7%) 16/43 (37.21%) 0.75  Proctitis 5/19 (26.3%) 14/24 (58.3%) 18/43 (41.9%) 0.04  Superbacterial infection 11/19 (57.9%) 13/24 (54.2%) 24/43 (55.8%) 0.81  Ophthalmologic complications 2/19 (10.5%) 1/24 (4.2%) 3/43 (7%) 0.58  Neuropsychiatric symptoms 4/19 (21.1%) 5/24 (20.8%) 9/43 (20.9%) >0.99  Odynophagia 6/19 (31.6%) 5/24 (20.8%) 11/43 (25.6%) 0.49 Need for invasive supportc 0/19 8/23 (34.8%) 8/42 (19.1%) 0.01 Urological or bowel obstruction 1/19 (5.3%) 3/23 (13%) 4/42 (9.5%) 0.61 Use of opioids for pain control 14/17 (82.4%) 13/23 (56.5%) 27/40 (67.5%) 0.09 Death 0/19 2/24 (8.3%) 2/43 (4.7%) 0.50

IQR, interquartile range.

an (%); median (IQR).

bFisher's exact test; Wilcoxon rank sum test; Pearson's chi-squared test.

cInvasive support included need for mechanical ventilation, vasoactive drugs, hemodialysis, venous or arterial lines and/or vesical catheterization.

When comparing to PWH and mpox who have not required inpatient care (n = 189), those hospitalized (n = 24) reported less frequent sex contact one month prior to symptoms onset (66.7 vs. 91.5%, P < 0.01) and presented worse outcomes related to the HIV care continuum, with poorer rates of retention (79.2 vs. 94.7%, P = 0.01), regular ART use (79.2 vs. 94.1%, P = 0.02) and virological suppression (66.7 vs. 87.7%, P = 0.01), in addition to showing lower CD4+ cell counts [median 448 (IQR 174–850) vs. 630 (IQR 487–889), P = 0.02] and less frequently presenting CD4+ counts higher than 350 cells/μl (58.3 vs. 91.8%, P < 0.01) (Table 3 and Fig. 1).

Table 3 - Sociodemographic, behavioral and clinical characteristics of mpox participants coinfected with HIV according to hospitalization during mpox follow-up (N = 213). Not hospitalized, (N = 189)a Hospitalized (N = 24)a Totala (N = 213) P valueb Median age (IQR) 35 (30–40) 35 (30–42) 35 (30–40) 0.75 Age range 0.83  18–24 years 11/189 (5.8%) 1/24 (4.2%) 12/213 (5.6%)  25–29 years 30/189 (15.9%) 3/24 (12.5%) 33/213 (15.5%)  30–39 years 93/189 (49.2%) 12/24 (50%) 105/213 (49.3%)  ≥40 years 55/189 (29.1%) 8/24 (33.3%) 63/213 (29.6%) Gender identity >0.99  Cisgender men 180/189 (95.2%) 24/24 (100%) 204/213 (95.8%)  Cisgender women 2/189 (1.1%) 0/24 2/213 (0.9%)  Travesti or TGW 7/189 (3.7%) 0/24 7/213 (3.3%) Race 0.75  White 55/154 (35.7%) 9/21 (42.9%) 64/175 (36.6%)  Indigenous 2/154 (1.3%) 0/21 2/175 (1.1%)  Pardo or mixed 52/154 (33.8%) 5/21 (23.8%) 57/175 (32.6%)  Black 45/154 (29.2%) 7/21 (33.3%) 52/175 (29.7%) Education 0.22  Primary 90/156 (57.7%) 10/24 (41.7%) 100/180 (55%)  Secondary 11/156 (7.1%) 1/24 (4.2%) 12/180 (7.6%)  Postsecondary 55/156 (35.3%) 13/24 (54.2%) 68/180 (38%) Reported sexual contact 30 days before symptoms onset 161/176 (91.5%) 14/21 (66.7%) 175/197 (88.8%) <0.01 Probably vaccinated for smallpox 15/189 (7.9%) 4/24 (16.7%) 19/213 (8.9%) 0.12 HIV-RNA viral load 0.04  <40 copies/ml 143/188 (76.1%) 13/24 (54.2%) 156/212 (73.6%)  41–200 copies/ml 21/188 (11.2%) 3/24 (12.5%) 24/212 (11.3%)  201–1000 copies/ml 6/188 (3.2%) 2/24 (8.3%) 8/212 (3.8%)  >1000 copies/ml 18/188 (9.5%) 6/24 (25%) 24/212 (11.3%) Median CD4+ (cells/μl, IQR) 630 (487, 889) 448 (174, 850) 624 (463, 889) 0.02 CD4+ range (cells/μl) <0.01