Retrospective analysis was used to analyze the information of the 169 chemotherapy patients with NSCLC and nosocomial infection during hospitalization; 170 patients without nosocomial infection were also involved as a control group. All the patients were diagnosed as non-small cell lung cancer by fiberoptic bronchoscopy and biopsy. All the patients in observation group had nosocomial infection. There were 126 males and 44 females in the observation group. The ages ranged from 45 to 70 with a mean age of 61±1.0 years. The duration of disease ranged from 3 months to 2 years with a mean value of 5.1±0.2 months. Other clinical data of the patients in observation group included length of hospitalization, invasive operation, combination of two or more medical diseases, the use of prophylactic antibiotics, use of glucocorticoids and immunosuppressive agents. The patients with immune system disease and the patients who were combined with malignant tumors were excluded. The study obtained approval from the Ethics Committee of Binzhou City Central Hospital and all the participants signed the informed consent before enrollment.

The distribution of pathogens in nosocomial infection in patients with advanced NSCLC was statistically analyzed. The distribution of major drug resistance of Gram-negative (G⁻) and Gram-positive (G⁺) were analyzed. The survival conditions of the patients were studied. The risk factors of nosocomial infection in patients with NSCLC were analyzed by univariate and multivariate logistic analyses.

According to the ‘National Clinical Laboratory Procedures’, routine testing methods were utilized. Drug susceptibility test was performed by paper diffusion method (K-B method). CLSI 2008–2010 operation procedures and result evaluation standards were used. Columbia blood plate, MH plates, identifying culture medium and drug sensitive slips were provided by the Oxoid (London, UK). The quality control strains viz. Staphylococcus aureus ATCC 25923, Pseudomonas aeruginosa ATCC 27853 and Escherichia coli ATCC 25922 were provided by provincial CDC clinical laboratory test center. The KB method is mainly used for the detection of ampicillin, cefotaxime, cefepime, ceftazidime, imipenem, clindamycin, vancomycin, erythromycin, roxithromycin, gentamicin and amikacin. The corresponding biological reagents were procured from Hangzhou Tianhe Microoragnism Reagents Co., Ltd. (Hangzhou, China). MH agar plates were stored at 4°C before use. Technicians with >5 years laboratory experience had performed all operations in strict accordance with the instructions.

The investigators were trained in a unified manner. The informed consent of the respondents was obtained. The investigators were familiar with the contents of the questionnaire and solutions to the frequently asked questions. Reliability coefficient was applied with values ranged from 0 to 1. The reliability of this study was 0.996. Two tests were performed using the same questionnaire on all subjects with an interval of 30 days to obtain the stability coefficient.

SPSS 21.0 software was utilized for statistical analyses (IBM, Armonk, NY, USA). The data were expressed as mean ± standard deviation. The t-test was used to compare the mean of the two groups. The χ² test was used to compare the rates of the two groups. The univariate and multivariate logistic regression analysis was used for the analysis of correlation between the factors and the disease. P<0.05 was considered as statistically significant.

G⁺ infection accounted for 45.6% cases. Staphylococcus aureus and coagulase-negative staphylococci were the most common species. The proportion of G⁻ infection was 54.4%, of which Pseudomonas aeruginosa and Klebsiella pneumoniae were most common (Table I).

The most common drug-resistant antibiotics for Pseudomonas aeruginosa were sulfamethoxazole, ciprofloxacin and cefotaxime, while the most common drug-resistant antibiotics for Klebsiella pneumoniae infection were sulfonamides and gentamicin (Table II).

The most common drug-resistant antibiotics for Staphylococcus aureus were penicillin G, ampicillin and sulfamethoxazole, while the most common drug-resistant antibiotics for coagulase-negative staphylococci were penicillin G, ampicillin and sulfamethoxazole. Both of the G⁺ bacteria were sensitive to vancomycin and teicoplanin, and even more sensitive to imipenem or meropenem (Table III).

Nosocomial infections were most commonly found in respiratory system followed by the urinary system and digestive system, infections in skin, soft tissue and other parts (such as intracranial infection and intra-articular infection) (Table IV).

The median survival time of the observation group was 5.1±0.3 months, while the median survival time of the control group was 8.5±0.6 months. Therefore, the median survival time of the observation group was significantly shorter than that of the control group (t=65.928, P<0.05) (Fig. 1). The median survival time of the observation group was 5.1±0.3 months, while the median survival time of the control group was 8.5±0.6 months. So, the median survival time of the observation group was significantly shorter than that of the control group (t=65.928, P<0.05).

Invasive operation, combination of two or more diseases, the use of glucocorticoids and immunosuppressive agents in the course of treatment were observed commonly for NSCLC patients who were older than 60 years and have been hospitalized for 30 days or more (Table V).

Multivariate regression analysis showed that age over 60 years, invasive operation, combination of two or more diseases, the use of glucocorticoids and immunosuppressive agents were independent risk factors for nosocomial infection in patients with advanced NSCLC (Table VI).