The association between serum CD4 T lymphocyte counts and surgical outcomes in HIV/AIDS patients in Guangxi, China: a retrospective cohort study

Introduction

Methods

Results

A total of 168 hospitalized adult HIV/AIDS patients who met our inclusion criteria and underwent surgery from 2016 to 2019 were included in this study. The method of inclusion in this study is continuous inclusion, which solves the selective bias. Three postoperative outcomes occurred in these patients: Postoperative complications occurred in 66 cases (39.29%). Poor healing was reported for Grade B or C incisions in 17 cases (10.12%). Moreover, two deaths related to surgical disease or surgical complications were recorded within 2 weeks after surgery (1.19%). The number of deaths did not meet the statistical requirements and were not included in this study’s outcome measures.

Table 1 illustrates patient demographics and surgical baseline data. The majority of patients in this cohort were males (76.8%), mainly young adults (72.0%) ranging from 18 to 60 years old, and farmers (66.7%) with at least a basic educational (74.4%) background. The HIV/AIDS infection route for the vast majority of patients was sexual transmission (94.6%) in Guangxi, China, a multi-ethnic region, with ethnic minorities (35.1%) making up roughly one-third of the patients. This is consistent with the demographic data of AIDS patients in China (National Health Commission of the people’s Republic of China, 2019). The number of patients with erythrocyte counts of less than 4.00/ml or 3.50/ml in men or women respectively at baseline was 54.8%, and the number of patients with a hemoglobin level of less than 120 g/l and 110 g/l in men and women, respectively, was 41.7%. Meanwhile, approximately 26.2% of these patients had preoperative comorbidities. Furthermore, 18.5% of patients were treated with ART preoperatively for less than 30 days, a statistically different finding between the two groups (P = 0.021). Other preoperative baseline values such as gender (P = 0.037), occupation (P = 0.002), preoperative hemoglobin level (P = 0.029), and plasma albumin level (P = 0.049) also elicited statistically significant differences between the two groups. In terms of the remaining demographic information and baseline values such as age, educational background, ethnicity, route of infection, BMI, erythrocyte counts, preoperative comorbidities and CD4/CD8 variables, none of the differences were statistically significant between the two groups (P ≥ 0.05).

Table 2 delineates the comparison of surgical characteristics between the two groups. With regard to the disease-specific subgroup analysis, orthopedic and joint surgery was more likely to be treated surgically (43.5%). Grade-III surgery with more complex procedures, higher technical requirements, and higher risks had the highest proportion in both groups (34.9% in the CD4-CC < 200 cells/μl group, and 46.4% in the CD4-CC ≥ 200 cells/μl group). It is also worth noting that these HIV/AIDS patients were more likely to undergo traditional surgical approaches (80.95%). The proportion of contaminated incisions in the CD4-CC < 200 group was significantly higher than that in the CD4-CC > 200 group and proved to be significantly different (P = 0.001). Other variables of surgical characteristics, such as the type of admission, type of disease, surgical grade, and minimally invasive surgery, did not display any statistically significant difference between the two groups.

Table 3 lists the risk-adjusted surgical outcomes. After adjusting for surgical baseline data such as gender, age, body mass index, erythrocyte counts, duration of ART, comorbidities and flow cytometry, as well as surgical characteristics including the type of surgery, type of disease, surgical procedure, surgical grade and incision classification, we found that the patients in the CD4-CC < 200 cells/μl group were 2.379 times more likely to experience postoperative complications (P = 0.038). The risk-adjusted P-value for poor incisional healing was lower in the CD4-CC < 200 cells/μl group, but the difference did not reach a statistical significance.

Table 4 demonstrates the single factor analysis stratification of the two groups. In the age stratification group, we found out that patients in the CD4-CC < 200 cells/μl group were 4.125-fold more likely to have postoperative complications than those in the CD4-CC ≥ 200 cells/μl group for patients older than 60 years old (P = 0.036). Erythrocyte counts stratification results revealed a 4.476-fold increase in the odds of postoperative complications ofthe CD4-CC < 200 cells/μl group in the subgroup below 4.00/ml in men or 3.50/ml in women (P = 0.002). Regarding emergency admissions, the odds of postoperative complications were 3.700 times higher in the CD4-CC < 200 cells/μl group than the CD4-CC ≥ 200 cells/μl group (P = 0.013). Stratified comparisons of hemoglobin levels and incisional classification were not statistically significant for the occurrence of postoperative complications. Similarly, stratified comparisons of age, erythrocyte counts, hemoglobin levels, incisional classification, and the type of admission were not statistically significant for incisional healing.

Table 5 presents a multifactorial analysis of patient stratification in both groups. After risk adjustment, the risk of complications in the subgroup older than 60 years with CD4-CC < 200 cells/μl was significantly increased compared with the univariate analysis (OR 27.504), which was statistically significant (P = 0.009). Additionally, the adjusted erythrocyte-count stratification indicated that the incidence of postoperative complications was still higher (OR 3.353) in the CD4-CC < 200 cells/μl group (P = 0.036) for both males and females. There was no statistically significant difference between emergency and elective surgeries after adjustment for stratified admission types.

Discussion

HIV/AIDS is now widely recognized as a chronic disease rather than an acute and life-threatening illness. In Guangxi, China, late detection of HIV/AIDS >50.00% (Ge et al., 2019) and treatment interruptions account for 4.8 per 100 persons a year of the infected cases (Liu et al., 2018), and CD4 counts are becoming increasingly necessary for the evaluation of surgical outcomes. In this study, three characteristics of the CD4-CC < 200 cells/μl group were highlighted and compared to those of the control (CD4-CC ≥ 200 cells/μl) group. After a thorough analysis of the collected results, gender and occupation displayed a significant difference between the two groups of patients. This difference may indicate that male and rural patients with CD4-CC < 200 cells/μl have a lower willingness to seek medical care, leading to late detection of HIV/AIDS or decreased adherence to ART treatment. Early detection and early treatment of HIV/AIDS has been mandated as critical by the Chinese National Health authorities. Furthermore, clinical attention should be paid to enhance knowledge of the disease in these populations to improve treatment compliance. Next, patients in the two cohorts underwent surgery with different indications. Compared to patients in the CD4-CC ≥ 200 cells/μl group, those in the CD4-CC < 200 cells/μl group differed from the baseline in terms of less than 30 days of HARRT, low erythrocyte counts, and low plasma albumin. Infectious diseases account for about two-thirds, whereas the surgical characteristics were shown in 65.1% of the CD4 < 200 cells/μl group regarding the proportion of contaminated or potentially contaminated incisions. This discrepancy points out a worse preoperative composite in patients with CD4-CC < 200 cells/μl. Additionally, studies on abdominal surgery and plastic surgery have documented that CD4-CC < 200 cells/μl may negatively affect postoperative outcomes (Deneve et al., 2010; Davison et al., 2008). To this end, the surgical outcome of this study was designed in two distinct phases: Postoperative complications, which reflect the perioperative period, and incisional healing, which reflects the final surgical outcome. The results revealed that adjusted postoperative complications were statistically different between the two groups; however, no statistically significant difference was identified with respect to poor incisional healing. In other words, there was a statistically significant difference in the perioperative period after surgery in the CD4-CC < 200 cells/μl group, but no difference in endpoint outcomes. Also, our study further analyzed stratification factors with subgroups such as the elderly and erythrocytopenic patients, which had a significantly increased risk of postoperative complications in the CD4-CC < 200 cells/μl group. Previous studies have not reported stratification factors. Our study reports that CD4-CC is not a definitive contraindication for surgery; however, the patient’s baseline and surgical characteristics may help predict surgical outcomes and allow for appropriate interventions and intensive care. This study was conducted among HIV/AIDS patients over the age of 18, so it is not suitable for teenagers and children on the generalizability of the results.

Previous studies have found that the rate of postoperative infection in orthopedic joint surgeries among HIV-positive patients was 23%, and CD4-CC < 300 cells/μl, hospitalization, multiple traumas, and low plasma albumin levels were all found to be associated with postoperative infections (Guild et al., 2012). In our study, osteoarticular surgery was indicated for up to 43.5% of HIV/AIDS patients with 17 infectious and noninfectious diseases, including ICD-10 codes A18.007 + M49.0*, M87.800 × 051, and S72.000; these codes correspond to Pot’s disease (tuberculosis spondylitis), ischemic necrosis of the femoral head, and femoral fracture, respectively. Epidemiological data from Guangxi indicated that the proportion of HIV/AIDS patients with concurrent extrapulmonary tuberculosis was 8.86% (Huang et al., 2020); meanwhile, AIDS unrelated diseases and fractures also accounted for a large proportion of these orthopedic surgery cases. As a result, the proportion of osteoarticular surgeries is higher in our data.

Tuberculosis (TB) is a disease closely linked to cellular immunity, and it remains a common opportunistic infection in HIV/AIDS patients in Guangxi, China. Despite the fact that the incidence of tuberculosis has dropped from 120/100,000 in 2016 to 90/100,000, it is still a common opportunistic infection in HIV/AIDS patients (From 11.5% to 32.5%) (Pang et al., 2018; Xiao et al., 2013). In this study, 34 patients were operated on for infectious diseases, of which 15 TB and extrapulmonary TB-related surgeries were performed, accounting for 44.12% of these surgeries. ICD-10 codes A16.201, A18.007 + M49.0*, A18.207, A18.205, A18.034 + M01.1*, A18.118 + N51.1* and A18.300 × 009 were included on the list of diseases. This may be indicative of a higher prevalence of Mycobacterium tuberculosis infection among HIV/AIDS patients. Consequently, the medical management of this population must include regular screening and preventive measures for latent infections. Tuberculosis screening in China has been included in the annual screening program for HIV/AIDS patients by the Chinese National Health Administration. As far as we know, the AIDS outpost hospital where these surgeries were performed was the only provincial-level outpost hospital in the region, and its capacity was far superior to that of the prefecture-level outpost hospital. Hence, the more complex and risky Class III surgeries were referred from lower-level hospitals, which led to an increase in the incidence of Class III surgeries in this population. The incidence of Class III surgeries had indeed increased, mostly due to referrals from lower-level hospitals rather than an inherent rise of Class III surgeries in these people.

This study has some limitations, mainly due to flaws in its retrospective design. Chinese patients with HIV/AIDS can receive a free annual viral load test provided by the Chinese government. In this study, 46.4% of patients underwent emergency surgeries, and even some of those who had elective surgeries might also have had viral load testing performed more than 3 months prior to surgery, making it less relevant to surgical outcomes. Therefore, although it would have been interesting to examine the impact of viral load on surgical outcomes, we chose to rely on CD4-CC to determine the participants’ HIV/AIDS status. Additionally, the HIS system is unable to track patients who are automatically discharged within 2 weeks after surgery, and a small number of cases with missing data may often be cases with a poor prognosis in which the patient or family members chose to abandon treatment. Therefore, these cases were not included in this study, which may affect the accuracy of the data.

Conclusions

The risk of surgical complications was significantly different in the CD4-CC < 200 cells/μl group compared to the CD4-CC ≥ 200 cells/μl group. Among the adjusted stratification factors, the risk of surgical complications was significantly increased in the elderly and erythrocytopenic patients in the CD4-CC < 200 cells/μl group, whereas no statistically significant difference was observed with respect to the risk of poorly healed incisions. A convincing explanation for this observation is that AIDS outpost hospitals have good diagnostic capabilities, and postoperative complications can be cured with medical intervention without affecting the final surgical outcome. This implies that CD4-CC are not a determinant factor in deciding to perform a surgical procedure but that the patients’ baseline and surgical characteristics may contribute to their surgical outcomes. Thus, appropriate interventions and intensive care should be administered when necessary. Notwithstanding, further studies are imperative to confirm these findings.

Supplemental Information