A comparison of COVID-19 inpatients by HIV status

Abstract

COVID-19 in-hospital morbidity and mortality in people living with HIV (PLWH) were compared to HIV-negative COVID-19 patients within a New York City metropolitan health system, the hardest hit region in the United States early in the pandemic. A total of 10,202 inpatients were diagnosed with COVID-19, of which 99 were PLWH. PLWH were younger (58.3 years (SD = 12.42) versus 64.32 years (SD = 16.77), p < 0.001) and had a higher prevalence of men (73.7% versus 57.9%, p = 0.002) and Blacks (43.4% versus 21.7%, p < 0.001) than the HIV-negative population. PLWH had a higher prevalence of malignancies (18% versus 7%, p = < 0.001), chronic liver disease (12% versus 3%, p < 0.001), and end-stage renal disease (11% versus 4%, p = 0.007). Use of a ventilator, admission to the ICU, and in-hospital mortality were not different. Of the 99 PLWH, 12 were virally unsuppressed and 9 had CD4% < 14. Two of the 12 virally unsuppressed patients and 4/9 patients with CD4% < 14 died. Ninety-one of the 99 PLWH were on treatment for HIV, and 5 of the 8 not on treatment died. Among PLWH with prior values, absolute CD4 count decreased an average of 192 cells/mm³ at the time of COVID-19 diagnosis (p < 0.001). Hospitalized patients with HIV and COVID-19 coinfection did not have worse outcomes than the general population. Among PLWH, those with CD4%<14 or not on treatment for HIV had higher mortality rates. Those PLWH who received IL-6 inhibitors had lower mortality rates. PLWH given antifungal medications, hydroxychloroquine, antibiotics (including azithromycin), steroids, and vasopressors had higher mortality rates.

Keywords

AIDS < viral disease human immunodeficiency virus < viral disease viral disease

As of January 2, 2021, 348, 411people have died of COVID-19 in the United States. SARS-CoV-2 is the causative agent behind the COVID-19 pandemic, and in some people, it leads to a dysregulated immune response, referred to as “cytokine storm.” Cytokine storm is associated with lymphopenia, a hypercoagulable state, and multiorgan disease.¹ Those patients with advanced age, male gender, immunosuppression, and higher comorbidity burden are considered at higher risk for severe COVID-19.²

Information is needed on how people living with HIV (PLWH) would fare with COVID-19. Initially, it was postulated that PLWH would have worse outcomes than HIV-negative patients due to their immune compromised status.³ Many PLWH share social determinants of health with communities with a high burden of COVID-19 infection. However, the spectrum of HIV-related immune suppression is broad, and published data have shown conflicting results in terms of PLWH and COVID-19 outcomes. Recently, a multicenter descriptive study showed that PLWH with a CD4 count <200 cells/mm³ had higher hospitalization rates, lower ICU-free survival, and decreased overall survival when compared to PLWH with CD4 counts >200 cells/mm³.⁴ In contrast, a study of PLWH reported that those who acquired COVID-19 infection had similar outcomes to HIV-negative patients in terms of severity of illness and mortality.⁵ This finding has been replicated in patient cohorts in other locations in the United States as well as internationally.^6-9 HIV is associated with immune dysregulation, not merely suppression. It is possible that the quality, rather than just the quantity, of the immune response may be a factor in the pathogenesis of the aberrant cytokine storm associated with COVID-19. The baseline immune activation associated with suppressed HIV may impact the immune response to infection with SARS-CoV-2.¹⁰ Persistent high levels of immune activation and correlations with HIV-1 proviral DNA and 2-LTR circle measures were observed in a cohort of Mexican individuals following long-term and fully suppressive treatment, further supporting the persistent inflammatory state associated with even treated HIV infection.¹⁰ Prior studies of SARS-CoV and MERS-CoV showed that PLWH did not contract SARS at the same rate nor progress to severe illness as often as the general population.^11,12 It was also thought that PLWH may be at lower risk for progression to severe disease because of their regular use of antiretrovirals.^12,13

The first case of the SARS-CoV-2 infection in New York occurred on March 1, 2020. Our health system cared for over 11,000 hospitalized patients with confirmed COVID-19 between March 2, 2020 and May 15 2020. We sought to evaluate the experience and outcomes of PLWH admitted to any of our health system hospitals with COVID-19 and compare these outcomes to the HIV-negative population of COVID-19 patients admitted within our health system during the same time period. We also aimed to evaluate differences within HIV and COVID-19 coinfected patients to determine whether outcomes differed based on viral suppression status.

Methods

Study design and sample

Data for this study were provided by the COVID-19 Research Consortium Registry from Northwell Health and electronic medical records (EMRs). Northwell Health consists of 23 hospitals across NYC, Long Island, and Westchester, encompassing an urban, suburban, and rural population region. The registry retrospectively collected demographic, lab, medication, and vital measures data from the EMR on all COVID-19 patients admitted to a hospital within Northwell Health between March 2, 2020, and April 27, 2020. Data indicating HIV status and other HIV-related variables were additionally extracted from the medical record. The Northwell Health Institutional Review Board approved this study as minimal-risk research and waived the requirement for informed consent.

All diagnosed COVID-19 patients admitted to a hospital within Northwell Health between March 2, 2020, and April 27, 2020, with complete discharge follow-up by June 30, 2020, were included. Patients <18 years of age or still admitted at the end of follow-up were excluded. HIV or AIDS diagnosis was indicated by ICD-10 codes. This study considered all COVID-19 inpatients in Northwell Health as a population and HIV inpatients with COVID-19 as a sample from that population. Given the low number of HIV COVID-19 inpatients (N = 99), this study was considered exploratory.

Measures

The primary outcome was length of hospitalization in days. Secondary outcomes included in-hospital mortality, ICU admission (yes/no), and intubation (yes/no).

Study variables included demographics (age in years, race, ethnicity, gender, language, and zip code percent of families living below poverty level), hospital, comorbidities, smoking status, lab measures, medication utilization, and vital sign measurements including temperature, blood pressures, respiratory rate, heart rate, and oxygen saturation. Febrile was defined as temperature greater than or equal to 100.5°F, and hypoxic was defined as oxygen saturation less than 94%. The first, average, minimum, and maximum lab values were extracted. The first vital measure from admission was extracted. HIV-related variables included HIV viral load, CD4 cell count at admission and within the last 6 months, and current ARV regimen.

Quantitative variables, such as age, vital measures, lab measures, and days of utilization, were evaluated for normality and if not normal, transformations were applied. Creatinine, CRP, D-dimer, ferritin, and lactate dehydrogenase values were transformed using the natural log.

Statistical methods

Patient characteristics were described overall and for the PLWH. For categorical variables, one-sample exact binomial tests were used to compare estimated proportions from PLWH to population proportions. For normally distributed continuous variables, one-sample t-tests were used to compare estimated means to population means. Estimates and 95% confidence intervals using exact methods are presented. Among PLWH, CD4 measures and viral suppression were compared by outcomes. CD4 measures at admission date (current) and within the past 6 months of the inpatient admission date (historic) were compared using paired t-tests for patients with both measures. Analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA).

Results

There were 10,202 patients identified through the Registry and the EMR that met inclusion criteria, of which 99 (0.97%) were PLWH. PLWH were significantly different from the general population by several factors (Table 1). PLWH were on average younger (58.3 years (SD = 12.42) versus 64.32 years (SD = 16.77), p < 0.001) and had a higher prevalence of men (73.7% versus 57.9%, p = 0.002) and Blacks (43.4% versus 21.7%, p < 0.001) than the general population. A greater proportion of PLWH resided in a zip code with 13%–47% of families living below poverty level (32% versus 26%, p < 0.001).

Table 1.

PLWH characteristics compared to the COVID-19+ inpatient population.

	Total (N = 10,202)		HIV patients (N = 99)			p value
Variable	n	%	n	%	95% CI
Age at admission, mean (SD)	64.32 (16.77)		58.3 (12.42)		(55.83–60.78)	<0.001
Gender, n %
Female	4294	42.09	26	26.26	(17.93–36.07)	0.002
Male	5908	57.91	73	73.74	(63.93–82.07)	0.002
Race, n %
Asian	867	8.5	1	1.01	(0.03–5.5)	0.003
Black	2212	21.68	43	43.43	(33.5–53.77)	<0.001
Other/multiracial	2855	27.98	27	27.27	(18.8–37.15)	0.977
White	3824	37.48	20	20.2	(3.55–15.3)	0.134
Unknown	444	4.35	8	8.08	(12.8–29.46)	<0.001
Ethnicity, n %
Hispanic or Latino	2222	21.78	23	23.23	(15.33–32.79)	0.803
Not Hispanic or Latino	7322	71.77	69	69.7	(59.65–78.53)	0.718
Other/unknown	658	6.45	7	7.07	(2.89–14.03)	0.915
Smoking, n %
Current/former	1983	19.44	25	25.25	(17.06–34.98)	0.188
Never	7481	73.33	65	65.66	(55.44–74.91)	0.112
Unknown	738	7.23	9	9.09	(4.24–16.56)	0.575
Insurance, n %
Commercial	2915	28.57	26	26..26	(17.93–36.07)	0.702
Medicaid	2231	21.87	32	32.32	(23.27–42.47)	0.021
Medicare	4774	46.79	36	36.36	(26.93–46.64)	0.047
Self-pay/other	282	2.76	5	5.05	(1.66–11.39)	0.279
% FPL, median (IQR)	6.93	4.79–13.75	10	4.96–18.45		<0.001

p values were based on one-sample binomial tests for categorical outcomes. p values to compare sample means to the population means were based on one-sample t-tests and to compare sample medians to population medians were based on Sign Tests.

SD = standard deviation; IQR = interquartile range; CI = confidence interval; FPL = families below poverty line; PLWH = people living with HIV.

PLWH also had a higher prevalence of diagnosis of malignancy (18.2% versus 7.5%, p < 0.001), chronic liver disease (12.1% versus 2.6%, p < 0.001), and end-stage renal disease (11.1% versus 4.3%, p = 0.007) (Figure 1). PLWH were found to have a lower prevalence of hypertension (44.4% versus 58.6) than the general population. There were no differences in prevalence of diabetes, asthma, heart failure, coronary artery disease, chronic obstructive pulmonary disease, chronic kidney disease, peripheral vascular disease, or obesity.

When comparing vital signs at time of admission, there were no differences in body mass index, systolic blood pressure, diastolic blood pressure, heart rate, or respiratory rate. PLWH were more often afebrile, defined as a temperature less than 100.5F (83.7% afebrile versus 75.3% afebrile), and less often hypoxic, defined as an oxygen saturation less than 94% (19.2% hypoxic versus 37.6% hypoxic, p < 0.001).

There were no differences found between PLWH and the general population in measurements of C-reactive protein, ferritin, lactate dehydrogenase, or platelet count. PLWH had significantly higher creatinine (1.15 mg/dL versus 0.99, p < 0.001) and D-dimer (1286 ng/mL versus 1041 ng/mL, p = 0.002) values on presentation (Table 2).

Table 2.

PLWH clinical characteristics compared to the COVID-19+ inpatient population.

Variable	Total (N = 10,202)		HIV patients (N = 99)			p value
Vitals	n	%	n	%	95% CI
BMI
<30	3982	61	45	72.58	(59.77–83.15)	0.078
≥30	2546	39	17	27.42	(16.85–40.23)	0.078
Diastolic BP
<60	1235	12.16	6	6.06	(2.26–12.73)	0.072
≥60	8925	87.84	93	93.94	(87.27–97.74)	0.072
Systolic BP
<90	317	3.12	3	3.03	(0.63–8.6)	0.002
≥90	9842	96.88	96	96.97	(91.4–99.37)	0.002
Heart rate
≤90	4148	40.75	50	50.51	(40.27–60.71)	<0.001
>90	6032	59.25	49	49.49	(39.29–59.73)	<0.001
Respiratory rate
≤20	6308	61.97	65	65.66	(55.44–74.91)	0.518
>20	3871	38.03	34	34.34	(25.09–44.56)	0.518
Oxygen saturation (%)
<94	3822	37.58	19	19.19	(11.97–28.34)	<0.001
≥94	6348	62.42	80	80.81	(71.66–88.03)	<0.001
Temperature (degrees F)
<100.5	7661	75.3	82	83.67	(74.84–90.37)	0.063
≥100.5	2513	24.7	16	16.33	(9.63–25.16)	0.063

Lab measures	Mean (SD)	Mean (SD)	95% CI
Creatinine (mg/dL), max	2.37 (2.78)	3.8 (5.31)	—	—
Log creatinine, max	0.47 (0.79)	0.78 (0.95)	(0.59–0.97)	<0.001
Maximum CRP (mg/L)	47.81 (83.91)	42.72 (73.46)	—	—
Log maximum CRP	2.55 (0.99)	2.54 (1.15)	(2.3–2.79)	0.136
Maximum D-dimer (ng/mL)	4614.32 (9143.67)	4858.97 (8634.16)	—	—
Log maximum D-dimer	7.26 (1.47)	7.48 (1.41)	(7.16–7.8)	0.002
Maximum ferritin (ng/mL)	2379.81 (5838.19)	2148.96 (3978.8)	—	—
Log maximum ferritin	7 (1.16)	6.97 (1.18)	(6.71–7.23)	0.872
Lactate dehydrogenase (U/L), max	599.67 (997.52)	569.15 (506.16)	—	—
Log lactate dehydrogenase, max	6.18 (0.55)	6.14 (0.6)	(5.99–6.28)	0.866
Maximum platelet count, K/uL	346.87 (151.59)	322.64 (141.23)	(294.33–350.96)	0.924
Minimum platelet count, K/uL	186.38 (84.67)	176.41 (93.14)	(157.73–195.08)	0.799

p values were based on one-sample binomial tests for categorical outcomes.

p values to compare sample means to the population means were based on one-sample t-tests and to compare sample medians to population medians were based on Sign Tests.

SD = standard deviation; IQR = interquartile range; CI = confidence interval; BP = blood pressure.

Lab measures based on extracting the maximum or minimum value within 48 h of admission.

Lab measures were compared using log-transformed values. Vital measures are first within 48 h of admission.

PLWH were given azithromycin (59.6% versus 48.8%, p = 0.006), antifungal medications (11.1% versus 4.1%, p = 0.005), and H2 antagonists (41.4% versus 23%, p < 0.001) more than the general population. There were no differences in rates of use of chloroquine, hydroxychloroquine, antibiotics aside from azithromycin, steroids, vasopressors, IL-1 inhibitors, and IL-6 inhibitors.

Use of a ventilator (20.2% versus 20.0%), admission to the ICU (25.3% versus 26.5%), and in-hospital mortality (25.3% versus 26.5%) were not different in PLWH compared to the population (Table 3). When stratified by comorbidity and medication use, PLWH with asthma, diabetes, and chronic liver disease had higher mortality rates than the population with these comorbidities. Those PLWH who received IL-6 inhibitors had lower mortality rates, but those on steroids and vasopressors had higher mortality rates compared to the population. Among those who died, PLWH were on average younger than those without HIV (64.3 versus 73.1 years).

Table 3.

Comparison of PLWH and HIV-negative inpatients with COVID-19 infection.

Outcome	Total (N = 10,202)		HIV patients (N = 99)			p value
	n	%	n	%	95% CI
Ventilation	2037	19.97	20	20.2	(12.8–29.46)	1.000
ICU	2699	26.46	25	25.25	(17.06–34.98)	0.888
Mortality	2703	26.49	25	25.25	(17.06–34.98)	0.883
LOS in days, median (IQR)	6.55 (3.28–12)		6.68 (2.94–12.2)			0.688

SD = standard deviation; IQR = interquartile range; CI = confidence interval; FPL = families below poverty line; PLWH = people living with HIV.

Of the 99 PLWH, 12 (12.1%) were virally unsuppressed and 9 (10.8%) had CD4% < 14. Two of the 12 (16.7%) virally unsuppressed patients died, and 4/9 (44.4%) patients with CD4% < 14 died. Mean absolute CD3 and CD4 counts, as well as percent, were slightly lower among those who died. Of the 25 PLWH who were admitted to the ICU, 20 were on a ventilator and 16 died. There were 8 patients not on ARVs, and of those, 3 (37.5%) died. Of the 91 patients on ARVs, 22 (24.2%) died (Table 4). Mortality rates were higher among PLWH with diabetes, chronic liver disease, and cancer. However, they were lower among PLWH with hypertension, heart failure, and ESRD than the general population with hypertension, heart failure, and ESRD. PLWH given antifungal medication, hydroxychloroquine, antibiotics (including azithromycin), steroids, and vasopressors had higher mortality rates. However, as described, those on ARVs had lower mortality rates.

Table 4.

Comparison of clinical characteristics by mortality among HIV patients (N = 99).

Variable	Total (N = 99)	Mortality = yes (n = 25)	Mortality = no (n = 74)	n ^a
Variable	Mean (SD)	Mean (SD)	Mean (SD)	n ^a
Age, years	58.3 (12.42)	64.32 (10.2)	56.27 (12.5)	99
Absolute CD4 count cells/mm³ at time of COVID-19 illness	333.76 (254.25)	313.04 (245.61)	341.53 (258.88)	88
Premorbid CD4 cells/mm³	537.18 (346.63)	627.11 (284.35)	514.06 (360.89)	44
CD4% at time of COVID-19 illness	27.37 (11.78)	25.09 (12.25)	28.25 (11.57)	83
Premorbid CD4%	27.88 (12.12)	27.22 (11.04)	28.06 (12.54)	43
	n (%)	n (%)	n (%)
Virally unsuppressed, n (%)	12 (12.12)	2 (16.67)	10 (83.33)	99
Virally suppressed, n (%)	87 (87.88)	23 (26.44)	64 (73.56)
CD4 pct <14, n (%)	9 (10.84)	4 (44.44)	5 (55.56)	83
CD4 pct ≥14, n (%)	74 (89.16)	19 (25.68)	55 (74.32)
ARV medication, n (%)				99
Yes	91 (91.92)	22 (24.18)	69 (75.82)
No	8 (8.08)	3 (37.5)	5 (62.5)

^aSome variables contained missing values. Number of non-missing values per variable presented.

Among PLWH with prior measurements (n = 44), the premorbid absolute CD4 count dropped by an average of 192 cells/mm³ when diagnosed with COVID (552 versus 360, p < 0.001). CD4 percent and CD4 to CD8 ratio were unchanged when comparing premorbid measurements with measurements taken during COVID illness. Amongst those PLWH who died, the average absolute CD4 count in the 6 months prior to COVID-19 infection was 627 cells/mm³, and the average absolute CD4 count at the time of COVID-19 diagnosis was 313 cells/mm³, showing a 50% decrease in absolute CD4. In those PLWH who recovered from COVID-19, the average absolute CD4 count was 514 cells/mm³ in the 6 months prior, and 341 cells/mm³ when measured during COVID-19 illness, a 34% decrease (Table 4).

Discussion

Overall, we describe a population of PLWH that is diverse, not only racially diverse, but diverse in terms of age, comorbidities, and socioeconomic status. The PLWH presented here are largely in care and virologically suppressed, a PLWH population quite different from other large health systems in the NYC area.^5,14 We found no differences in mortality and severe outcomes in hospitalized COVID infected PLWH when compared to the general COVID-19 admitted population in our health system. The COVID-19 patients admitted to the hospital likely reflect a sicker segment of the overall COVID-19 cases in the general NY or US population. PLWH required vasopressors and mechanical ventilation at a rate similar to the general COVID-19 population. Our findings mirror findings from prior studies comparing the outcomes of PLWH to HIV-negative COVID patients.⁵

There were some notable differences between the general hospitalized COVID patient population and PLWH who were hospitalized with COVID that can potentially be accounted for by the location of this study. PLWH were younger, more likely to be male, and more likely to identify as Black. This observed higher percentage of PLWH who were male and identify as Black mirrors known demographics of the PLWH population in the NYC metropolitan area. Specifically, PLWH in our study were more likely to be receiving public assistance, Medicare, or Medicaid than the larger COVID infected population. Nearly one-third of the HIV patients lived in a zip code with 13–47% of families below poverty level, as compared to only roughly a quarter of the HIV-negative population.¹⁵ This is reflective of New York City’s hotspots in April 2020 being characterized as middle-class working neighborhoods, or neighborhoods where many “essential workers” reside.¹⁶ Further, these differences in demographics generally reflect the demographics of PLWH. It is important to note that both HIV and COVID-19 disproportionately affect Black and Brown Americans.¹⁷ The out of proportion spread of COVID-19 in people of color in the United States has been associated with poverty, living in densely populated households, and areas with high air pollution.¹⁷

Despite their younger age, PLWH had multiple comorbidities, consistent with the accelerated/accentuated aging associated with HIV infection. In this highly treated and virally suppressed population, the long-term effects of the chronic inflammation and ARV toxicities associated with treated HIV infection are expected.¹⁸ PLWH had a significantly higher prevalence of cancer, which is currently the leading cause of death in PLWH in care.¹⁸ Chronic liver disease and end-stage renal disease were also more common in the PLWH, both of which may be due to chronic inflammatory changes, and coinfection with viral hepatitis. Given that the PLWH had higher rates of grim comorbidities, it is remarkable that there were no observed differences in mortality. Because of the premature aging and chronic immune activation associated with HIV infection, it is not surprising that our HIV patient population, despite their younger age, had similar rates of CAD, PVD, and COPD to the older general population of COVID-19 infected patients. It is possible that the increased prevalence of chronic comorbidity associated with HIV immune activation contributed to the “age gap” observed between PLWH with COVID and the overall population regarding illness severity and mortality. It would be important to identify which aspects of immune dysfunction play a role in this pathogenic process. It is intriguing that our data demonstrate that use of IL-6 antagonists was associated with a better response in PLWH compared to the general population. The SMART study found that elevated IL-6 levels were associated with morbidity and mortality in PLWH.¹⁹ As such, it is possible that elevated baseline IL-6 in PLWH may be associated with enhanced COVID disease responses. IL-6 levels were not routinely measured so could not be included in this report; however, this presents an area for further study.

PLWH are diverse and can present with a broad range of clinical states, from young, healthy, virologically suppressed, and immune intact patients to those with multiple comorbidities and advanced immune suppression. When comparing the HIV-positive patients who died to those who survived COVID-19, there were no apparent differences in CD4:CD8 ratio, CD4 percent, or absolute CD4 count at the time of illness. This differs from a prior, smaller study done in the Bronx which showed patients with lower CD4 counts had higher mortality rates¹⁴ although our small sample size limits our ability to make a strong conclusion about our finding. It is possible that the population of hospitalized PLWH with COVID in the Bronx may have had a higher proportion of patients with AIDS, or that absolute CD4 measured after admission with acute COVID-19 may be less reliable as a measure of immune system status. We examined changes in CD4 count among those who had pre-COVID measures accessible in the 6 months prior to COVID-19 diagnosis. We found an average drop in absolute CD4 of 192 cells/mm³, a 35% decrease. A study conducted in Wuhan, China, examined T cell counts in HIV-negative patients with COVID-19 and found that the values were suppressed to approximately half of the normal values.²⁰ When compared by severity of illness, those patients who were critically ill were found to have significantly lower CD3, CD4, and CD8 counts compared to those patients considered severely ill and mildly ill.²⁰ Our findings were similar among the PLWH in that those patients who died from COVID-19 had a large decrease in their CD4 counts compared to those PLWH who recovered.

To further explore how the immune effects of HIV may contribute to outcomes related to symptomatic COVID infection, those with CD4 percentage below 14% (those with current AIDS status) were compared with those with CD4 measures above 14%. While absolute CD4 counts fluctuate based on several factors, CD4% tends to remain more stable. Nine patients of the 99 HIV positive met criteria for AIDS with CD4 percent less than 14. Four out of the 9 patients (44%) with CD4 <14% died. This finding, if replicable in a larger population, may indicate that the prevalent state of immune balance, rather than the current absolute CD4 count, may be more predictive of poor outcome among PLWH with COVID requiring hospitalization.

PLWH who were not virologically suppressed were found to have a higher mortality than the general population. Our findings are different from a multicenter study in New York City that did not find a higher death rate in non-virologically suppressed patients.⁵ This difference may be attributed to the small numbers and cannot be applied to the HIV population without further studies.

In those HIV patients who were not on ARVs, there was a higher likelihood for ICU admission, mechanical ventilation, and mortality compared to patients who were on treatment for HIV. These findings support the results of a study completed in China, reviewing 20 COVID-19 positive HIV patients found that patients on ARVs prior to COVID infection had less severe inflammatory responses.²⁰ It is not clear currently whether the ARV medications themselves influence the progression of COVID-19, or if it is the immunologic benefits the patients receive from HIV viral suppression that led to better outcomes.²¹

It is important to note those PLWH who were hospitalized for COVID-19 infection were less often febrile and hypoxic when compared to the overall COVID-19 infected population, suggesting that PLWH were not as severely ill. The authors attribute this to a lower threshold for admission for PLWH in our hospital system. This may also be reflected in the higher use of azithromycin, antifungals, and H2 blockers (as a potential COVID therapy) among PLWH.

Of note, three patients (3%) were newly diagnosed with HIV during their admission for COVID-19. Two of the patients accepted universal screening in the ED. One patient was screened after an employee’s needlestick injury. New diagnosis of HIV at the time of COVID-19 diagnosis is not commonly reported at this time but brings forward the importance of considering additional diagnoses in patients with COVID-19 that are critically ill²² and to emphasize the need for universal offering of HIV testing at the time of hospital admission.

Our study has limitations, mainly due to our small sample size of COVID hospitalized PLWH. PLWH in the study were generally virologically controlled and may not be reflective of sicker, poorly controlled HIV populations. The CD4 and viral load was measured only on a subset of patients admitted for COVID-19, making the sample sizes for these analyses too small to detect significant differences. Larger studies are needed to understand the clinical course of SARS-CoV-2 in PLWH and to elucidate optimal treatment plans. Further studies may also evaluate the clinical implications of a drop in CD4 count as an indicator of severity of COVID-19 illness.

This new epidemic that we collectively face emphasizes both the advances that have been made in HIV care and the dire need to end the HIV epidemic. We are now finding that when PLWH are treated and achieve an immunologic reconstitution, they mirror the outcomes of the HIV-negative population with some notable exceptions. The coalescence of the HIV and COVID-19 epidemics also demonstrates how important it is to identify new HIV diagnoses and to engage those newly diagnosed in care as these patients may be at risk for worse COVID-19 outcomes. It is important to point out that the majority of the PLWH with COVID infection in this study did not receive their routine HIV care within our health system. In future studies, we plan to identify the rate of SARS-CoV-2 infection in those PLWH who are engaged in care within our system.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Sarah Flannery

References

Giannis

Ziogas

Gianni

. Coagulation disorders in coronavirus infected patients: COVID-19, SARS-CoV-1, MERS-CoV and lessons from the past. J Clin Virol 2020; 127: 104362.

Espinosa

Zanetti

ADS

Antunes

, et al. Prevalence of comorbidities in patients and mortality cases affected by SARS-CoV2: a systematic review and meta-analysis. Rev Inst Med Trop Sao Paulo 2020; 62: e43. DOI: 10.1590/S1678-9946202062043.

Zhou

, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study The LancetS0140673620305663), 101016/S0140-673620. Lancet 2020; 395(10229): 1054–1062.

Dandachi

Geiger

Montgomery

, et al. Characteristics, comorbidities, and outcomes in a multicenter registry of patients with HIV and coronavirus disease-19. Clin Infect Dis. DOI: 10.1093/cid/ciaa1339.Epub ahead of print 2020

Sigel

Swartz

Golden

, et al. Coronavirus 2019 and people living with human immunodeficiency virus: outcomes for hospitalized patients in New York City. Clin Infect Dis 2020; 71(11): 2933–2938. DOI: 10.1093/cid/ciaa880.

Childs

Post

Norcross

, et al. Hospitalized patients With COVID-19 and human immunodeficiency virus: a case series. Clin Infect Dis 2020; 71(8): 2021–2022. DOI: 10.1093/cid/ciaa657.

Härter

Spinner

Roider

, et al. COVID-19 in people living with human immunodeficiency virus: a case series of 33 patients. Infection 2020; 48: 681–686.

Vizcarra

Pérez-Elías

Quereda

, et al. Description of COVID-19 in HIV-infected individuals: a single-centre, prospective cohort. Lancet HIV 2020; 7: e554–e564.

Isernia

Julia

Le Gac

, et al. SARS-COV2 infection in 30 HIV-infected patients followed-up in a French University Hospital. Int J Infect Dis 2020; 101: 49–51.

10.

Orta-Resendiz

Viveros-Rogel

Fuentes-Romero

, et al. Persistent high levels of immune activation and correlations with HIV-1 proviral DNA and 2-LTR circle loads in a cohort of Mexican individuals following long-term and fully suppressive treatment. Int J Infect Dis 2020; 100: 184–192. DOI: 10.1016/j.ijid.2020.08.044.

11.

Peeri

Shrestha

Rahman

, et al. The SARS, MERS and novel coronavirus (COVID-19) epidemics, the newest and biggest global health threats: what lessons have we learned?. Int J Epidemiol 2020; 49: 717–726.

12.

Zhang

Wang

. COVID-19 in people with HIV. Lancet HIV 2020; 7: e524–e526.

13.

Laurence

. Why aren’t people living with HIV at higher risk for developing severe coronavirus disease 2019 (COVID-19)? AIDS Patient Care STDS 2020; 34: 247–248.

14.

Suwanwongse

Shabarek

. Clinical features and outcome of HIV/SARS-CoV-2 coinfected patients in The Bronx, New York city. J Med Virol 2020; 92: 2387–2389.

15.

Team

. Zip Codes With The Highest Percentage Of Families Below Poverty Level In New York | Zip Atlas. [online] Zipatlas.com. Available at: http://zipatlas.com/us/ny/zip-code-comparison/families-below-poverty-level.htm (2020, September 2020).

16.

Maroko

Nash

Pavilonis

. COVID-19 and Inequity: a comparative spatial analysis of New York City and Chicago hot spots. J Urban Health 2020; 97: 461–470.

17.

Millett

Honermann

Jones

, et al. White counties stand apart: the primacy of residential segregation in COVID-19 and HIV diagnoses. AIDS Patient Care STDS 2020; 34: 417–424.

18.

Althoff

Smit

Reiss

, et al. HIV and ageing: Improving quantity and quality of life. Current Opinion in HIV and AIDS 2016; 11(5): 527–536. DOI: 10.1097/COH.0000000000000305.

19.

Grund

Baker

Deeks

, et al. Relevance of interleukin-6 and D-dimer for serious non-AIDS morbidity and death among HIV-positive adults on suppressive antiretroviral therapy. PLoS One 2016; 11: e0155100.

20.

FAN

WANG

, et al. Suppressed T cell-mediated immunity in patients with COVID-19: a clinical retrospective study in Wuhan, China. J Infect 2020; 81(1): e51–e60. DOI: 10.1016/j.jinf.2020.04.012.

21.

Liu

Zeng

Cao

, et al. Effect of a previous history of antiretroviral treatment on the clinical picture of patients with co-infection of SARS-CoV-2 and HIV: a preliminary study. Int J Infect Dis 2020; 100: 141–148. DOI: 10.1016/j.ijid.2020.08.045.

22.

Zhang

Ding

, et al. New HIV diagnoses in patients with COVID-19: two case reports and a brief literature review. BMC Infect Dis 2020; 20: 771–810.