Clinical characterization and placental pathology of mpox infection in hospitalized patients in the Democratic Republic of the Congo

Abstract

We describe the results of a prospective observational study of the clinical natural history of human monkeypox (mpox) virus (MPXV) infections at the remote L'Hopital General de Reference de Kole (Kole hospital), the rainforest of the Congo River basin of the Democratic Republic of the Congo (DRC) from March 2007 until August 2011. The research was conducted jointly by the Institute National de Recherche Biomedical (INRB) and the US Army Medical Research Institute of Infectious Diseases (USAMRIID). The Kole hospital was one of the two previous WHO Mpox study sites (1981-1986). The hospital is staffed by a Spanish Order of Catholic Nuns from La Congregation Des Soeurs Missionnaires Du Christ Jesus including two Spanish physicians, who were members of the Order as well, were part of the WHO study on human mpox. Of 244 patients admitted with a clinical diagnosis of MPXV infection, 216 were positive in both the Pan-Orthopox and MPXV specific PCR. The cardinal observations of these 216 patients are summarized in this report. There were three deaths (3/216) among these hospitalized patients; fetal death occurred in 3 of 4 patients who were pregnant at admission, with the placenta of one fetus demonstrating prominent MPXV infection of the chorionic villi. The most common complaints were rash (96.8%), malaise (85.2%), sore throat (78.2%), and lymphadenopathy/adenopathy (57.4%). The most common physical exam findings were mpox rash (99.5%) and lymphadenopathy (98.6%). The single patient without the classic mpox rash had been previously vaccinated against smallpox. Age group of less than 5 years had the highest lesion count. Primary household cases tended to have higher lesion counts than secondary or later same household cases. Of the 216 patients, 200 were tested for IgM & IgG antibodies (Abs) to Orthopoxviruses. All 200 patients had anti-orthopoxvirus IgG Abs; whereas 189/200 were positive for IgM. Patients with hypoalbuminemia had a high risk of severe disease. Patients with fatal disease had higher maximum geometric mean values than survivors for the following variables, respectively: viral DNA in blood (DNAemia); maximum lesion count; day of admission mean AST and ALT.

Fig 1. The mean of duration only included clinical symptoms and clinical signs present on the admission day.

Fig 2. The figure provides a graphical representation of relative lesion density for dermatological regions.

Each dot notionally represents approximately 10 lesions per body region as a function of the region’s relative surface area (mean of lesion count/percentage of surface area).

Fig 3. The mean with standard error bars of total lesion count by age group.

(A) Repeated measurement Analysis of Variance (ANOVA) showed there was no significant difference among the groups (p = 0.2552). (B) Distribution of mean of lesion counts for all patients by body location over time.

Fig 4. Lesion progression from macule, papule, vesicle, pustule, umbilication (Umbilic), scabbing (scab) and desquamation (desquam).

◊ represent mean of count, ─ represent median of count, ° represent outliers, ↓ represent median day after onset of rash to get max lesion count for that specific rash morphology on hand.

Fig 5. Mpox associated lymph node distribution. The lymph node count and distribution were documented on the day patients had the most lymph nodes.

One dot equals 10 lymph nodes or fraction thereof depending upon the count.

Fig 6. Mean of PCR count from blood, throat and scab (Log₁₀) with standard error bars (Unit: genomes/mL) over time.

Fig 7

A. Maternal surface of the placenta with multiple grossly visible areas of basal hemorrhage. B. Fetal skin with multifocal positivity for virus using immunohistochemistry. Antibody to vaccinia virus counterstained with hematoxylin & eosin, X10. C. In the placenta immunohistochemistry demonstrates a pattern of intense and diffuse staining of chorionic villus stromal cells for MPXV that is consistent with macrophages (Hofbauer cells). The infected cells are increased in number, termed Hofbauer cell hyperplasia, a pattern that is seen in some TORCH viral infections. Antibody to vaccinia virus counterstained with hematoxylin & eosin, x4.

Fig 8. Forest Plot showing statistically significant associations.

Odd ratio (OR), upper 95% confidence interval (UCL), and lower 95% confidence interval (LCL) were calculated using separate generalized estimating equations (GEE) with cumulative logit models, day after rash onset was adjusted as covariate. Adjusted p value were calculated using stepdown Bonferroni correction if needed. Only significant results are showed here (p value ≤ 0.05).