Changes in HIV-Related Hospitalizations During the HAART Era in an Inner-City Hospital

The implementation of highly active antiretroviral therapy has been associated with decreased hospitalizations among HIV-infected patients. We previously reported our experience with hospitalizations in the HAART era and noted a modest decline in hospitalizations as well as continued high mortality among patients admitted with severe opportunistic infections.^1,2 In this study, we evaluated trends among HIV-infected patients admitted to our hospital over a 6-year period, from 2000 to 2005.

METHODS
Stroger Hospital of Cook County is a 470-bed hospital that predominantly cares for the inner-city poor in the Chicago and Greater Chicago area. The hospital has had an HIV unit or service since 1990, and case report forms have been kept on all admissions to the HIV unit from September 1999 through June 2001. After June 2001, case report forms were completed on all HIV-related admissions in the hospital.

HIV Unit Patient Care ProtocolsCase report forms. Physician assistants, nurse practitioners, or research assistants completed case report forms during admission. Information was collected on patient demographics, HIV risk factors, diagnoses leading to hospitalization, outpatient clinic attendance, substance abuse, antiretroviral therapy, opportunistic infection prophylaxis before admission, and HIV RNA level and CD4 count. Data were also abstracted from medical and pharmacy records and the hospital information system. Primary diagnoses were determined by one of the authors (JP) after reviewing the completed case report form. Diagnoses leading to hospitalization were divided into the following categories: HIV-related, other infectious disease–related, internal medicine–related, or other.

Definition of terms. HIV-related admission was defined as admission for infections or malignancies related to an immunocompromised state. These included HIV-related opportunistic infections and malignancies as well as AIDS-related complex symptoms and signs. Specific diagnosis was based on opportunistic infection (eg, Cryptococcus infection), malignancy (eg, Kaposi sarcoma), or HIV-related syndrome (eg, aphthous stomatitis). Infectious disease admissions were admissions for infections that were not AIDS-related, and the specific diagnoses were based on the organ system infected or specific infectious organism. Internal medicine admissions were admissions for internal medicine–related issues that were not AIDS-related, and specific admissions were based on the organ system involved (eg, cardiovascular, GI). “Other admissions” is a miscellaneous category that was broadly divided into admissions for psychological reasons, substance abuse issues, adverse drug events, traumas and fractures, pregnancy, and social issues. It should be noted that most women admitted for pregnancy were hospitalized outside the HIV unit and that these admissions were not captured until after June 2001.

Hepatitis C virus infection was defined as a positive antibody test (enzyme-linked immunoassay followed by recombinant immunoblot assay for confirmation) before or during study admission. HIV viral load and CD4 counts were evaluated from 1 year before study admission to 1 month after study admission, with the values closest to the study admission date recorded.

Highly active antiretroviral therapy was defined as the concurrent receipt of at least 3 antiretrovirals. Patients were defined as receiving highly active antiretroviral therapy if they reported taking an antiretroviral regimen consistent with highly active antiretroviral therapy on admission to the hospital. Substance abuse (both licit and illicit) before admission was defined as the self-reported active use of illicit drugs or excessive use of alcohol at the time of admission as determined by the admitting service. Homelessness was self-reported by patients and defined as living on the streets or in a shelter. HIV risk factors were self-reported by patients and included sex between men, injection drug use, heterosexual sex, vertical transmission, and blood transfusion. New patients were those not previously seen in the outpatient HIV clinic nor previously admitted to Cook County Hospital.

Statistics
All data were analyzed using SPSS 10.1 (SPSS, Inc, Chicago). Quality checks were performed on the database to ensure accuracy of inputted values. Descriptive statistics were used to evaluate demographics. Nominal variables were evaluated using chi-square, ordinal variables by Kendall correlations, and continuous variables by analysis of variance and t tests. All nominal data were presented by P value, odds ratio, and 95% confidence intervals. Ordinal variable data were presented by P values. Univariate variables with a P value of .10 or less were evaluated for multivariate predictors by backward linear regression analysis, with removal of the variable if the P value exceeded .06.

The following analyses were also done: analysis based on year of admission, comparison of demographic factors between African American patients and non–African American patients, comparison between patients older than 50 years on admission and those 50 years old or younger, and a comparison of admission causes between patients with and without injection drug use as an HIV risk factor.

RESULTSDemographics

During the 6-year study period, there were 6045 admissions, with significant demographic changes in our population over the years (Table 1). Of note was the decline in admissions of substance abusers, illicit substance abusers, and patients who gave injection drug use as their HIV risk factor. Associated with the above was a decrease in admissions of patients coinfected with hepatitis C or B virus as well as homeless patients.

There were changes regarding ethnicity in our population, with a significant decline in admission of African Americans and rise in admission of Asian patients, with a trend of increasing admissions among Latino and white patients. This was also seen in overall hospital and clinic visits (HIV-infected and noninfected patients) to the hospital from 2000 to 2005 (Figure 1). There was a significantly higher incidence of substance abuse (75% vs 51%; P < .001), illicit substance abuse (65% vs 36%; P < .001), injection drug use as an HIV risk factor (44% vs 22%; P < .001), and homelessness (11.5% vs 4.9%; P < .001) among the HIV-infected African American patients. It should also be noted that injection drug use was associated with a significantly higher incidence of hepatitis C (63.1% vs 19.6%; P < .001), home-lessness (15.8% vs 6.4%; P < .001), absence of highly active antiretroviral therapy on admission (38.8% vs 47.7%; P < .001), and admissions for infectious disease reasons (47.5% vs 40.5%; P < .001) and hepatic reasons (2.9% vs 1.6%; P < .001).

Figure 1. Patient visits (HIV-infected and non–HIV-infected) to Stroger Hospital of Cook County from 2000 to 2005 by race.

Although the mean age of our patient population did not change, the percentage of patients older than 50 years increased, indicating the aging of our patient population (P < .001). In an analysis comparing reasons for admission between patients aged 50 years or younger and those older than 50, patients older than 50 were more likely to be admitted for internal medicine–related reasons (48.6% vs 25.5%; P < .001) and less likely to be admitted for HIV-related reasons (24.6% vs 39.6%; P < .001).

The mean CD4 count and viral load among our patients seesawed over the years. The mean and median CD4 count increased over the last 2 years of the study from a nadir in 2003, which inversely correlated with the mean and median viral load. These changes may reflect the relative change in the number of patients receiving highly active antiretroviral therapy on admission to the hospital, with a nadir of 37.6% in 2003 and a subsequent rise to 48.9% in 2005.

Diagnosis Leading to Hospitalization
There was a significant increase in the number of admissions for internal medicine–related, HIV-related, and other diagnoses (all P < .05); however, the increase in HIV-related diagnoses was not as great as those in the other two categories (Figure 2).

Figure 2.Patient admissions to Stroger Hospital of Cook County from 2000 to 2005 by category of admission. (ID, infectious disease; IM, internal medicine.)

Among HIV-related diagnoses (Figure 3), there was significant rise in the number of patients admitted with lymphoma and oral/esoph­ageal candidiasis and a significant decline in admissions involving infection with Cryptosporidium, Mycobacterium kansasii, Mycobacterium avium complex, and Aspergillus. These changes may reflect a continued high prevalence of early opportunistic infections and a decline in traditionally late opportunistic infections.

Figure 3.Changes in specific HIV-related admissions.

Among infectious disease–related admissions (Figure 4), there has been no change in the total number of non–HIV-associated infectious diseases. However, there have been striking changes in the types of infections reported. Since 2002, there was a decline in admitted patients with upper respiratory tract infections. This may reflect changes in admissions since the new hospital opened in December 2002. Because of the smaller number of beds in the new hospital, patients who present to the emergency department with non-severe symptoms are monitored in a 23-hour observation area of the emergency department and may not be admitted. This area was not available in the old hospital, and these patients were more likely admitted. Therefore, patients presenting with an upper respiratory tract infection were less likely to be admitted in the new hospital. There has also been a significant rise in skin and soft tissue infections among our population. This appears to be coincidental with the outbreak of community-acquired methicillin resis tant Staphylococcus aureus (CA-MRSA) infections we have been seeing in our hospital.

Figure 4.Changes in infectious disease–associated admissions. (URI, upper respiratory tract infection; UTI, urinary tract infection.)

As noted earlier, there has been a rise in admissions with internal medicine diagnoses. Specifically, we have seen a rise in admissions for renal and endocrine diagnoses as well as a rise in admissions for non–HIV-associated malignancies. There has been a decline in admissions for hematological and hepatic reasons (Figure 5). The rise in renal-related admissions occurred despite a drop in admissions among African Americans. Yet, in the African American population, there was a significant increase in admissions for renal issues vers

us the rest of the population (3.8% vs 1.9%; P = .001) The decline in hematological admissions was not associated with a decline in the use of zidovudine (Table 2). The decline in hepatic admissions likely reflects the decrease in admissions of patients coinfected with hep­atitis B or C virus and the decline in the number of patients who had a history of injection drug use.

Figure 5. Changes in admissions for specific internal medicine diagnoses.

The increase in admissions for other reasons may partially be an artifact of data collection. As stated in the methods section, case report forms were completed only on patients admitted to the HIV unit up to June 2001.

Pregnant HIV-infected women tended to be admitted to the obstetrics service and were not included in our database until after June 2001. Their addition to the database after June 2001 can mostly explain the rise in other admissions. However, it should be noted that there was also a rise in psychiatric admissions and adverse reactions to medications (Figure 6).

Figure 6.Changes in admissions for “other” diagnoses.HIV Medications
As noted in Table 1, there was no significant change in the overall use of highly active antiretroviral therapy. However, there have been dramatic changes in the medications used, including significant declines in the utilization of stavudine, indinavir, nel­finavir, saquinivir, and nevirapine and an increased utilization of abac­avir, tenofovir, efavirenz, lopinavir/ ritonavir, and newer agents such as atazanavir, fosamprenavir, and em­tricitabine (Table 2). These changes may reflect movement away from older antiretroviral agents because of short- or long-term toxicity or because newer coformulations facilitate medication adherence by decreasing pill burden and frequency.

DISCUSSION
Our study evaluated admission patterns at our hospital as the HAART era progressed. As noted in the results section, there have been significant changes in the demographics, reasons for admission, and use of antiretroviral agents among our inpatient population. These changes reflect the complex interactions between socioeconomic factors affecting our inpatient population and the impact of HIV therapies.

The main demographic changes include changes in patient ethnicity, a decrease in the prevalence of active substance abuse among our inpatients, and a decline in the number of patients coinfected with hepatitis C virus and of those with injection drug use as their HIV risk factor. These changes may be explained by events external to the HIV epidemic. These include a grow­ing immigrant population (Hispanic and sub-Saharan African) in the Chicagoland area; real estate development of formerly low-income property, which has forc­ed many poor African Americans to live in areas away from the hospital; a new hospital facility; and loss of or inadequate medical insurance among other HIV-infected populations not traditionally admitted to the Stroger Hospital of Cook County.³

The construction of the new hospital was completed in December 2002, and the hospital served as an attractant to a population that pre­viously was not admitted to Cook County Hospital. This has included a higher proportion of admissions among Latino, white, and sub-Saharan African patients to the HIV ser­vice. This change has also been seen among the general hospital popu­lation, with even more dramatic changes in ethnicity than occurred on the HIV service.⁴ The gentrifi­cation of parts of the city that pre­viously were occupied by low-income housing projects has displaced many of the former residents (mostly African Americans) to collar areas of the city or to border suburbs distant from the central location of the county hospital. The displacement of the urban poor (among whom there was a high incidence of substance abuse) is probably par­tially, if not fully, responsible for the change in the HIV risk factors of our group and can explain the reduction in the incidence of substance abuse, illicit substance abuse, injection drug use as an HIV risk factor, hepatitis C, and homelessness in our population.

Although many authors have noted a decline in opportunistic infections in the HAART era,^5,6 there are little data on the incidence of opportunistic infections as the HAART era has progressed. Grubb and associates⁷ noted a continued drop in the diagnosis of Pneumocystis pneumonia in a comparison of HIV Outpatient Study data from 1996-1999 and 2000-2003. Pallela and coworkers,⁸ working from the same data set, noted a rise in deaths related to hepatic causes and non–AIDS-associated malignancies. Some authors have also noted that survival was significantly shorter in patients who began highly active antiretroviral therapy with a CD4⁺ cell count of less than 200/µL.^9,10 Wood and colleagues¹¹ also noted the association of a high initial viral load and lack of adherence to antiretroviral therapy with mortality and the development of opportunistic infection in patients with baseline CD4⁺ cell counts above 200/µL. Sackoff and associates¹² re­viewed death certificates from AIDS patients in New York City and reported that from 1999 to 2004 there was a rise in deaths associated with non–HIV-associated causes. They noted an increase in deaths related to substance abuse, cardiovascular causes, and non–HIV-associated malignancies.

Our study showed no change in the overall incidence of patients admitted for HIV-associated reasons. However, we did see a difference in the types of opportunistic infections these patients had. This change may suggest that more often patients are being admitted with earlier opportunistic infections and less often with later opportunistic infections.¹³Pneumocystis jiroveci (formerly carinii) pneumonia and oral/esophageal candidiasis tend to occur earlier in the spectrum of HIV-associated op­por­tunistic infections than M kansasii, M avium complex, Crypto­sporidium, and Aspergillus infections, which suggests that manifestations of far advanced HIV disease are becoming less frequent. Although opportu­nistic infections continue to be a problem, patients are less likely to have progression to more advanced HIV stages that bring with them later-onset opportunistic infections. This may reflect the impact of highly active antiretroviral therapy.

We have also noted a significant increase in the number of admissions for lymphoma. Some authors have described a decline in HIV-asso­ci­ated lymphomas in the HAART era,^14,15 while others have noted a rise in malignancies in HIV-infected patients during the HAART era.^15,16 Our study also showed a rise in non–HIV-associated malignancies, as was described else­where.^8,12 Patel and coworkers¹⁷ noted a rise in lung cancer, Hodgkin lymphoma, anorectal cancer, and melanoma in HIV-infected patients dur­ing the HAART era, while Cadranel and associates¹⁸ noted that lung cancer in HIV-infected patients tended to occur at a younger age and with a more advanced presentation of disease associated with a poorer prognosis. Bini and colleagues¹⁹ noted a higher incidence of neoplastic lesions in the distal colon of HIV-infected patients older than 50 years compared with a control population older than 50 years.

Over the years, we noted a higher proportion of admitted patients older than 50 years. This may explain the rise in internal medicine admissions in our population. With the suc­cessful use of highly active antiret­roviral therapy, many HIV-infected patients are living much longer and are therefore prone to illnesses associated with aging.²⁰ Some of these illnesses may be accelerated in HIV-infected patients as a result of the HIV infection itself or the associated antiretroviral therapy. Some studies have documented the development of certain internal medicine diseases in a younger-than-expected population.^21,22 The premature development of internal medicine problems that are usually seen in older patients, as well as the higher incidence of ma­lignancies, may be associated with incomplete immune reconstitution with antiretroviral therapy.

Besides admissions for cancer-related issues, other specific internal medicine admissions that increased during this time include admissions for renal and endocrine problems. The rise in admissions for renal issues was substantial. Renal pathology in HIV-infected patients is common. Its most noted form is HIV-associated nephropathy,²³ which is seen more frequently in African Americans.²⁴ It is interesting that in our study, the number of African American admissions decreased in the last years of the study while admissions for renal issues increased.

Although there was no significant change in the percentage of infectious disease admissions, there were significant rises in skin and soft tissue infections and syphilis-related admissions. The rise in admissions for skin and soft tissue infections is coincidental with a rise in such infections in our hospital related to CA-MRSA.²⁵ The rise in admissions for syphilis in our patient population is a disturbing trend and suggests more unprotected high-risk sexual contacts among our patients. A rise in sexually transmitted disease also has been noted by other authors.²⁶

With progression of the HAART era, we have seen a change in the antiretroviral agents that our patients were taking. This was likely motivated by simpler regimens with lower pill burden, less frequent administration, and fewer side effects. This change in antiretroviral therapy has made it easier for patients to be adherent to therapy and may partially explain why more patients are admitted with lower viral loads. However, the overall usage of highly active antiretroviral therapy among our patients remains remarkably low (about half the population), even in patients with lower CD4 counts, which may explain the continued high number of admissions who have HIV-related problems. This low usage occurred despite the fact that there are a number of HIV clinics affiliated with the hospital and its health care bureau that provide antiretroviral therapy regardless of patients’ ability to pay.

As mentioned in the methods section, only patients admitted to the HIV unit were included in our study population from January 2000 until June 2001. After June 2001, all HIV-related admissions in the hospital were evaluated. This may have skewed some of the data, especially admissions for pregnancy-related issues. Other populations that may have been underrepresented before June 2001 include patients admitted to the surgical service, as well as patients directly admitted to the ICU, who did not survive to be transferred to the HIV unit. Patients with a new HIV diagnosis who were admitted to the general medicine units may not have been transferred to the HIV unit during hospitalization and would not have been recorded before June 2001.

Other limitations to our study were the retrospective nature of the evaluation, that it only looked at admission patterns at one hospital, and that the impact of outside factors may have significantly affected the admission patterns and demographics of our study population. However, our hospital is a typical inner-city urban hospital, and these findings may be consistent with findings in similar hospitals in the United States and perhaps in other developed countries. Also, demographic changes in the inner city may reflect the ongoing real estate boom and immigration patterns, which also likely affect other inner-city hospitals.

In conclusion, the demographics in our inpatient population have changed significantly. These changes may be reflective of locally related as well as HIV-related issues. However, there has been a significant change in the reasons for admission among HIV-infected patients, with more patients admitted with internal medicine problems. These changes challenge HIV specialists caring for hospitalized HIV-infected patients to be well-versed in internal medicine as well as HIV-related diagnoses.

Dr Pulvirenti reports serving as a speaker for Bristol-Myers Squibb and for GlaxoSmithKline. No other potential conflict of interest relevant to this article was reported by the authors.

References:

References

1.

Pulvirenti J, Glowacki R, Muppidi U, et al. Hospitalized HIV-infected patients in the HAART era: a view from the inner city.

AIDS Patient Care STDs

. 2003;17:565-574.

2.

Pulvirenti J, Glowacki R, Herrera P, et al. High mortality in urban substance abuse patients after hospitalization for severe opportunistic infections.

2004 Infectious Disease Society of America

. Boston. Abstract 04-AB-1567.

3.

Johnson KM. Demographic Trends in the Chicago Metropolitan Area. Available at: www.luc.edu/depts/sociology/johnson/ChicagoCensus2004.html. Accessed July 6, 2007.

4.

Stroger Hospital Information Services. Hospital and Clinic Visits 2000-2005 by Race.

5.

Pallela FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators.

N Engl J Med

. 1998;338:853-860.

6.

Paul S, Gilbert HM, Ziecheck W, et al. The impact of potent antiretroviral therapy on the characteristics of hospitalized patients with HIV infection.

AIDS

. 1999;13:415-418.

7.

Grubb JR, Moorman AC, Baker RK, Masur H. The changing spectrum of pulmonary disease in patients with HIV infection on antiretroviral therapy.

AIDS

. 2006;20:1095-1107.

8.

Pallela FJ, Baker RK, Moorman AC, et al. Mortality in the highly active antiretroviral era: changing causes of death and disease in the HIV Outpatient Study.

J Acquir Immune Defic Syndr

. 2006;43:27-34.

9.

Moore DM, Hogg RS, Chan K, et al. Disease progression in patients with virological suppression in response to HAART is associated with the degree of immunological response.

AIDS

. 2006;20:371-377.

10.

Chadborn T, Baster K, Delpech V, et al. No time to wait: how many HIV-infected homosexual men are diagnosed late and consequently die? (England and Wales, 1993-2002).

AIDS

. 2005;19:513-520.

11.

Wood E, Hogg RS, Yip B, et al. Impact of baseline viral load and adherence on survival of HIV infected adults with baselines CD4 cell counts > or = 200 cells/microl.

AIDS

. 2006;20:1117-1123.

12.

Sackoff JE, Hanna DB, Pfeiffer MR, Torian LV. Causes of death among persons with AIDS in the era of highly active antiretroviral therapy: New York City.

Ann Intern Med

. 2006;145:397-406.

13.

Hoover DR, Saah AJ, Bacellar H, et al. Clinical manifestations of AIDS in the era of Pneumocystis prophylaxis. Multicenter AIDS Cohort Study.

N Engl J Med

. 1993;329:1922-1926.

14.

Hishima T, Oyaizu N, Fugii T, et al. Decrease in Epstein-Barr virus-positive AIDS-related lymphoma in the era of highly active antiretroviral therapy.

Microbes Infect

. 2006;8:1301-1307.

15.

Bower M, Palmieri C, Dhillon T. AIDS-related malignancies: changing epidemiology and the impact of highly active antiretroviral therapy.

Curr Opin Infect Dis

. 2006;19(1):14-19.

16.

Bonnet F, Lewden C, May, D, et al. Malignancies-related causes of death of HIV-infected patients in the era of highly active antiretroviral therapy. 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco. Abstract 120.

17.

Patel P, Novak RM, Tong T, et al. Incidence of non-AIDS defining malignancies in the HIV Outpatient Study. 11th Conference on Retroviruses and Opportunistic Infections; February 8-11, 2004; San Francisco. Abstract 81.

18.

Cadranel J, Garfield D, Lavolé A, et al. Lung cancer in HIV infected patients: facts, questions and challenges.

Thorax

. 2006;61:1000-1008.

19.

Bini EJ, Park J, Francois F. Use of flexible sigmoid­oscopy to screen for colorectal cancer in HIV infected patients 50 years of age and older.

Arch Intern Med

. 2006;166:1626-1631.

20.

Shah SS, McGowan JP, Smith C, et al. Comorbid conditions, treatment, and health maintenance in older persons with human immunodeficiency virus infection in New York City.

Clin Infect Dis

. 2002;35:1238-1243.

21.

Varriale P, Saravi G, Hernandez E, Carbon F. Acute myocardial infarction in patients infected with human immunodeficiency virus.

Am Heart J

. 2004;147:55-59.

22.

Burke M, Furman A, Hoffman M, et al. Lung cancer in patients with HIV infection: is it AIDS-related?

HIV Med

. 2004;5:110-114.

23.

Humphreys MH. Human immunodeficiency virus-associated glomerulosclerosis.

Kidney Int

. 1995;48:311-320.

24

. Rao TKS. A decade of human immunodeficiency virus-associated nephropathy (HIVAN).

Transplant Proc

. 1993;25:2439-2440.

25.

Popovich K, Weinstein R, Rice T, et al. Are community-associated methicillin resistant

Staphylococcus aureus

(CA-MRSA) strains replacing traditional nosocomial MRSA? 17th Annual Meeting of the Society for Healthcare Epidemiology of America; April 14-17, 2007; Baltimore. Abstract 162.

26.

Buchacz K, Greenberg A, Onorato I, Janssen R. Syphilis epidemics and human immunodeficiency virus (HIV) incidence among men who have sex with men in the United States: implications for HIV prevention.

Sex Transm Dis

. 2005;32(10 suppl):S73-S79.