Introduction

Infection during hospitalization, especially pulmonary infection, is a serious complication observed among patients with ST-segment elevation myocardial infarction (STEMI), which induces a tenfold increase in the 30-day mortality rate [1,2,3,4]. Early identification and intervention improve the effectiveness of treatment and outcomes.

Stress-induced hyperglycemia (SIH), defined as temporarily increased acute blood glycemia during an emergency situation, is a common state among patients with STEMI [5]. SIH is linked to infection both in vitro and in vivo [6, 7], and results in an increased release of proinflammatory cytokines [8, 9]. Prior studies have used admission blood glucose (ABG) as a parameter of stress hyperglycemia [10, 11]. However, the ABG level is influenced not only by acute stress but also by a chronic glycemic condition, which limits its ability to distinguish a true acute glycemic rise [12]. In recent years, the stress hyperglycemia ratio (SHR), calculated from the ABG adjusted by the average glycemic status [13], has been regarded as a better index of relative SIH compared to ABG alone [12]. The SHR has been reported as a reliable predictor for poor long- and short-term prognosis in patients with STEMI [12, 14, 15], and for stroke-associated pneumonia in patients without diabetes mellitus (DM) [16]. However, the prognostic value of the SHR for pulmonary infection during hospitalization among patients with STEMI remains unclear.

Therefore, the main aim of the present study was to investigate the association between SHR and pulmonary infection during hospitalization among patients with STEMI.

Method

Study population

Patients with STEMI at Guangdong Provincial People’s Hospital were consecutively enrolled between 2010 and 2020. Patients were diagnosed with STEMI according to the international guideline as follows: (1) typical chest pain or ischemia symptoms; (2) dynamic changes in the new electrocardiogram: ST-segment elevation of more than two adjacent leads or new left bundle branch block; (3) increased levels of myocardial-injury biochemical markers [17]. We excluded patients who (1) undergoing coronary artery bypass grafting, (2) with infection before the diagnosis of STEMI, (3) were on hemodialysis at admission, (4) died within 24 h after admission, (5) readmission to hospital, (6) did not receive the percutaneous coronary intervention (PCI) and (7) were missing crucial laboratory data (ABG or glycosylated hemoglobin [HbA1c]). This study was approved by the research ethics committee of Guangdong Provincial People’s Hospital and was conducted according to the guidelines stipulated in the Declaration of Helsinki.

Data collection and definitions

Baseline blood samples were collected within 24 h of admission and tested for white blood cell count, hemoglobin, ABG, HbA1c, troponin I/T, cardiac enzymes, serum creatinine, blood lipids, and other routine blood tests. Relevant data were collected, including demographic information, medical history (hypertension, DM, myocardial infarction [MI] or coronary revascularization, stroke), admission status (heart rate, systolic and diastolic blood pressure, and Killip class), and coronary revascularization data. Killip class is used to assess the severity of acute myocardial infarction. Killip class I was defined as absence of congestive heart failure, class II as presence of rales and/or jugular venous distension, class III as presence of pulmonary oedema and class IV as cardiogenic shock. DM was defined based on the history of DM or newly diagnosed DM with HbA1c ≥ 6.5%, fasting blood glucose ≥ 7.0 mmol/L, or 2-h plasma glucose ≥ 11.1 mmol/L in the oral glucose tolerance test [18]. The SHR was defined as the index calculated by the formula: [(ABG (mg/dl))/(28.7 × HbA1c (%) − 46.7)] [13].

Coronary angiography and/or PCI were performed at the interventional cardiologists’ discretion and according to the current guidelines [19]. Medications, including aspirin, clopidogrel or ticagrelor, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, and β-blockers, were prescribed by the cardiologists according to current guidelines and recommendations. Transthoracic echocardiography, electrocardiograph, and X-ray were performed, and chest computed tomography was performed if necessary.

Endpoints and definitions

The primary endpoint was pulmonary infection during hospitalization, which was determined based on typical medical imaging and clinical signs, symptoms or relevant laboratory biomarkers (e.g., white blood cell count) by an experienced clinician [20]. For patients without sufficient evidence recorded for pulmonary infection diagnosis, the infection was determined using ICD-10-CM codes at discharge. The secondary endpoint was in-hospital major adverse cardiovascular events (MACEs), comprising all-cause mortality, stroke, target vessel revascularization, or recurrent MI.

Statistical analyses

Continuous variables are shown as the mean ± SD or median (interquartile range) and were compared using independent t-test or Kruskal–Wallis test based on their distribution. Categorical variables are presented as numbers (percentages), and differences between groups were analyzed using Pearson’s chi-squared test or Fisher’s exact test. Variables with a statistical significance level of P < 0.10 on univariate logistic regression or with clinical importance, including age, sex, anemia, estimated glomerular filtration rate (eGFR), current smoker, DM, hypertension, chronic obstructive pulmonary disease, prior MI, prior PCI, prior stroke, transradial assessment, multi-vessel disease, and white blood cell count, were included and used for adjustment in the multivariate analysis. Receiver operating characteristic (ROC) curves were used to show the predictive value of the SHR. The best cutoff value and area under the ROC curve (AUC) were calculated. A cubic spine model was also performed to judge the effect of SHR on pulmonary infection. The logistic regression model was used to construct the cubic splines, and the knots were set at 0.75, 1, 1.25, 1.5, mainly based on the distribution of SHR. Statistical analysis was performed with SAS 9.4 (SAS Institute, Cary, NC). P < 0.05 were considered to indicate statistical significance, and all analyses were two-tailed.

Results

Baseline characteristics

From 2010 to 2020, 4,757 patients were diagnosed with STEMI. After excluding 42 patients undergoing coronary artery bypass grafting, 68 with infection before the diagnosis of STEMI, 25 on hemodialysis at admission, 49 subjects who died within 24 h after admission, 87 readmissions to hospital, 554 subjects who did not receive PCI and 1,091 without ABG or HbA1c, 2,841 patients were finally included. The study flow is shown in Fig. 1. The patients were divided into three groups according to the tertiles of SHR (n = 947, 946 and 948). The patients’ baseline characteristics are shown in Table 1.

Fig. 1
figure 1

Flow chart of participants. STEMI ST-segment elevation myocardial infarction, CABG coronary artery bypass grafting, PCI percutaneous coronary intervention, ABG admission blood glucose, HbA1c glycosylated hemoglobin, SHR stress hyperglycemia ratio

Full size image
Table 1 Comparison of baseline characteristics between groups stratified by tertile of SHR
Full size table

The mean age of the population was 62.27 ± 12.18 years, and 82.6% were male. 737 (83.6%) of the 882 DM diagnosis were based on medical history. For the remanent 145 newly diagnosis of DM, 15 (10.3%) participants were diagnosed with DM based on oral glucose tolerance test, 109 (75.2%) based both on glycemia and HbA1c, 13 (9.0%) based on HbA1c and 8 (5.5%) on glycemia only. As shown in Table 1, the higher SHR group was more likely to be older, female, has a higher Killip classification, and has a history of DM, but less likely to be current smokers. Meanwhile, the patients with lower SHR had a higher rate of prior MI and PCI, and in terms of in-hospital medication usage, a higher rate of aspirin use.

SHR for pulmonary infection and in-hospital MACEs

During hospitalization, 323 (11.4%) patients developed the pulmonary infection and 165 (5.8%) suffered from in-hospital MACEs. The SHR was associated with pulmonary infection (8.1%, 9.9%, and 18.0%; P < 0.001) and in-hospital MACEs (3.7%, 5.1%, and 8.6%; P < 0.001), as shown in Fig. 2. In multivariate logistic regression analysis, SHR, as a continuous value, remained an independent risk factor significantly associated with pulmonary infection during hospitalization (adjusted odds ratio [OR] = 1.46, 95% confidence interval [CI] 1.06–2.02, P = 0.021) and in-hospital MACEs (adjusted OR = 1.67, 95% CI 1.17–2.39, P = 0.005) after adjusting for potential risk factors, including age, sex, anemia, eGFR, current smoker, DM and so on (Table 2). Taking SHR as a categorical value, the same results were found, in that a SHR more than 1.176 was a risk factor for a higher risk of pulmonary infection during hospitalization (adjusted OR = 1.75, 95% CI 1.25–2.46, P = 0.001) and in-hospital MACEs (adjusted OR = 1.68, 95% CI 1.09–2.60, P = 0.020) (Table 2).

Fig. 2
figure 2

Incidence of in-hospital events in different SHR groups. ***P ≤ 0.001. SHR stress hyperglycemia ratio, MACEs major adverse cardiovascular events

Full size image
Table 2 Multivariable logistic regression analysis for the SHR as categorical variable and continuous variable
Full size table

Cubic spline models demonstrated no significant non-linear relationship between SHR and pulmonary infection during hospitalization (P = 0.210) or MACEs (P = 0.743) (Fig. 3).

Fig. 3
figure 3

A Cubic spine models for the association between SHR and pulmonary infection. B Cubic spine models for the association between SHR and in-hospital MACEs. SHR stress hyperglycemia ratio, MACEs major adverse cardiovascular events

Full size image

For the interaction between DM and the SHR, significant interactions were observed both for the pulmonary infection (interaction P = 0.012) and the in-hospital MACEs (interaction P = 0.009). For further analysis, we divided these patients into 6 groups, G1: without DM, SHR < 0.929 (T1); G2: without DM, 0.929 ≤ SHR < 1.176 (T2); G3: without DM, SHR ≥ 1.176 (T3); G4: with DM, T1; G5: with DM, T2 and G6: with DM, T3.

Multivariate logistic regression analysis demonstrated that DM was not significantly associated with the pulmonary infection (OR = 1.60, 95% CI 0.90–2.83, P = 0.110) nor the in-hospital MACEs (adjusted OR = 1.05, 95% CI 0.49–2.26, P = 0.898) for patients without stress hyperglycemia (T1). A SHR more than 1.176 (T3) was significantly associated with a higher risk of pulmonary infection during hospitalization (adjusted OR = 2.01, 95% CI 1.31–3.07, P = 0.001) but not for in-hospital MACEs (adjusted OR = 1.36, 95% CI 0.78–2.38, P = 0.276) for patients without DM.

A cumulative effect was observed for the in-hospital MACEs since DM with high SHR (T3) associated with higher risk than that of DM with low SHR (T1) and high SHR (T3) without DM. No cumulative effect was observed for the outcome of pulmonary infection. (Additional file 1: Table S1).

Predictive value of the SHR for in-hospital events

The ROC curve of the SHR for pulmonary infection during hospitalization was drawn, and, as shown in Fig. 4A, had an AUC of 0.624 (95% CI 0.590–0.659). The best cutoff value of 1.073 was derived with a sensitivity of 64.7% and specificity of 56.8%. Subgroup analysis showed that the AUC was 0.595 (95% CI 0.538–0.652) and 0.637 (95% CI 0.593–0.680) for patients with and without DM, respectively (Fig. 4B). Additionally, an AUC of 0.629 (95% CI 0.584–0.674) was found for in-hospital MACEs (Fig. 4C).

Fig. 4
figure 4

A ROC curve of SHR for pulmonary infection among STEMI patients. B ROC curve of SHR for pulmonary infection among STEMI patients with or without DM. C ROC curve of SHR for in-hospital MACEs among STEMI patients. ROC curve receiver operating characteristic curve, AUC area under the ROC curve, DM diabetes mellitus, SHR stress hyperglycemia ratio, STEMI ST-segment elevation myocardial infarction, MACEs major adverse cardiovascular events

Full size image

Discussion

To the best of our knowledge, this is the first study to establish a correlation between SHR and pulmonary infection during hospitalization among patients with STEMI undergoing PCI. Our main findings were that the SHR is independently correlated with pulmonary infection and MACEs in patients with STEMI during hospitalization.

SIH and STEMI

SIH is a common status with a strong predictive value for worse clinical outcomes for patients with STEMI [5, 10, 11]. Several complex acute physiological changes, such as excessive gluconeogenesis, activation of the sympathetic-adrenergic-system, counter-regulatory hormones (e.g., catecholamine, cortisol), and proinflammatory cytokines (e.g., tumor necrosis factor-α ), play roles in the development of SIH and lead to a vicious cycle [8, 9, 21, 22]. This vicious cycle is also believed to contribute to worse outcomes for critically ill patients, including all-cause mortality and pneumonia [10, 11, 16, 23]. The relationship between SIH and the development of long- and short-term MACEs in patients with STEMI has been proven by prior studies [12, 14, 15]. In this research, we drew the same conclusion that SHR, reflecting SIH, is independently associated with the risk of MACEs during hospitalization. Despite firm evidence of the relationship between SIH and MACEs, the correlation between SIH and pulmonary infection during hospitalization for patients with STEMI remains unclear.

SHR with pulmonary infection in patients with STEMI

As an important complication of STEMI, infection during hospitalization is associated with a worse clinical outcome [1,2,3,4], and several risk factors [24, 25] and scores [4, 26, 27] are utilized for infection prediction and prevention. Unfortunately, these predictive tools are largely complex or unable to be interfered.

In this research, we used the SHR, a simple and readily available parameter, on behalf of SIH, and found that it is independently associated with the risk of pulmonary infection during hospitalization for patients with STEMI. Although the underlying mechanism of this finding remains unclear, we hypothesize that the high pulmonary infection rate among high SHR patients with STEMI is attributed to the harmful effects of stress hyperglycemia. Glucose is considered a pro-inflammatory mediator, and elevation of blood glucose not only induces the generation of inflammatory cytokines, inflammatory processes, and insulin resistance [8, 9, 21], but also decreases vascular endothelial nitric oxide, which leads to vasoconstriction and low organ perfusion [28]. Moreover, glucose directly inhibits the function of T lymphocytes, immunoglobulin, and complement, disrupts the immune system, and further increases the risk of pulmonary infection [28]. In diabetic animal models, hyperglycemia can also lead to lower bacterial clearance and a higher rate of infection-related mortality [29,30,31]. Moreover, patients with high SHR have been found to be more likely to suffer from in-hospital MACEs [12, 32,33,34], which may increase the usage of invasive procedures, such as mechanical ventilation, thereby increasing the risk of pulmonary infection [35, 36]. This result provides a potential standard for admission blood glucose control for patients with STEMI during the acute phase, although further studies are needed.

This research found no significant non-linear relationship between the SHR and pulmonary infection or MACEs during hospitalization. Similarly, in the research of Yang et al. [34], although a J-shape relationship was confirmed between the SHR and major adverse cardiovascular and cerebrovascular events, MACE, cardiac death, and MI at a 2-year follow-up, neither a U nor J-shape relationship was found for in-hospital cardiac death and MI. These consistent findings highlight that the influence of the SHR on different outcomes in different periods may vary and have diverse mechanisms, which cannot yet be fully explained. Further studies are needed to clarify this phenomenon.

SHR with infection in patients with STEMI with or without DM

The SHR was also found to be independently associated with the risk of pulmonary infection during hospitalization among patients with STEMI with or without DM. In the ROC curve of SHR for pulmonary infection, the AUC of patients without DM was slightly larger than that of patients with DM (P = 0.051). This tendency is consistent with previous studies that focused on other aspects of STEMI outcomes [12, 14]. As an explanation, chronic hyperglycemia of DM may induce antioxidant defenses and protect tissues and cells from the oxidative stress of acute hyperglycemia [37,38,39].

Although a strong association between SIH and STEMI prognosis has been declared, the optimal management for SIH is still debated. Moreover, clinical trials of SIH therapies have yielded conflicting results. Indeed, Kosiborod et al. [40] showed an association between glucose normalization and better survival in patients with acute myocardial infarction and SIH. In contrast, a meta-analysis of three clinical studies revealed few benefits of intensive blood glucose control, but an increased risk of serious hypoglycemia in patients with acute myocardial infarction and DM [41]. In this research, the SHR was proven to be independently associated with the risk of pulmonary infection and MACEs during hospitalization for patients with STEMI undergoing PCI. As the SHR, a better index of relative stress hyperglycemia, was shown to be a better predictor of worse outcomes, we propose that glycemic targets based on SHR instead of ABG may be applied to the management of acute hyperglycemia in further studies.

Strengths and limitations

To the best of our knowledge, this is the first study to establish the correlation between SHR and pulmonary infection during hospitalization among patients with STEMI who underwent PCI. However, our study has several limitations. First, as this is an observational study, we could not adjust for all potential confounders. Second, only patients with STEMI in one center were included, which may limit the generalization of our conclusion. Third, blood glucose levels are prone to fluctuation, and the variation in durations from the onset of symptoms to the measurement of ABG and SHR may have impacted the results. Fourth, because of the small sample size of patients with STEMI without PCI, the impact of SHR on outcomes in these patients remained unclear. Therefore, further studies are needed to confirm the association between the SHR and pulmonary infection during hospitalization in patients with STEMI without PCI treatment.

Conclusion

The SHR is independently associated with the risk of pulmonary infection during hospitalization and in-hospital MACEs for patients with STEMI who underwent PCI.