The NIS database is the largest fully paid inpatient database in USA. As part of the Health Care Cost and Utilization Project (HCUP) and the Institute for Health Care Research (IHR), including general, short-term, non-federal and other specialty hospitals, NIS includes a stratified sample of discharges from over 1000 hospitals throughout USA. The provided sample represents an estimated 20% of the total yearly hospitalizations in the country (Rhemtulla et al. 2021). The data provided includes sampling weights used to calculate prevalence estimates for diseases across the country. Data can be extracted from the database, encompassing patient profiles, service levels, overall hospital expenditures, and diagnostic codes in accordance with the International Classification of Diseases, Ninth Revision, Clinical Modification ([ICD-9-CM] and ICD-10-CM) clinical modification.

The diagnostic and procedural information used in this study was derived from the ICD-9-CM and ICD-10-CM coding systems. This study aimed to conduct a comprehensive data query of the NIS database, focusing on main TSA cases that occurred between 2010 and 2019. Specifically, instances where the primary procedure code recorded were 81.80/81.81/81.88/81.97 (ICD-9-CM) and 0RRJ00Z/0RRJ07Z/0RRJ0J6/0RRJ0J7/0RRJ0JZ/0RRJ0KZ/0RRK00Z/0RRK07Z/0RRK0J6/0RRK0J7/0RRK0JZ/0RRK0KZ (ICD-10-CM) were examined. The exclusion criteria were as follows: (1) age < 18 years; (2) admission method: emergency, urgent, neonatal or another type of emergency room admission; and (3) pathological fractures of the humerus and clavicle.

Pulmonary complications were operationally defined in this study as.

(1) pneumonia (ICD-9-CM: 480/480.0/480.1/480.2/480.3/480.8/480.9/481/482/482.0/482.1/482.2/482.3/482.30/482.31/482.32/482.39/482.4/482.41/482.42/482.49/482.8/482.81/482.82/482.83/482.84/482.89/482.9/483/483.0/483.1/483.8/484/484.1/484.3/484.5/484.6/484.7/484.8/485/486;ICD-10-CM:J12/J12.0/J12.1/J12.2/J12.3/J12.8/J12.89/J12.9/J13/J14/J15/J15.1/J15.2/J15.20/J15.21/J15.211/J15.212/J15.29/J15.3/J15.4/J15.5/J15.6/J15.7/J15.8/J15.9/J16.0/J16.8/J17/J18/J18.0/J18.1/J18.2/J18.8/J18.9), (2) respiratory failure (ICD-9-CM: 51851, 51853, 51884, 51883, 51881, 77084; ICD-10-CM: J9582, J95821, j95822, J9600, j9602, J969, J961, J9601, J960, J962, J9610, J9612, J9620, P285, J96, J9611, J9691, J9690, J9692, J9621, J9622) and (3) PE (ICD-9-CM: 415.11, 415.19; ICD-10-CM: I260, I2699). (4) The term ‘any pulmonary complication’ was used to refer to the existence of one or more of the aforementioned issues affecting the pulmonary system. Patient characteristics, specifically age, gender and race, were studied. The analysis included outcome markers such as the mode of admission, type of insurance, length of stay (LOS), total cost of hospital stay, preoperative comorbidities and in-hospital mortality. ICD-9-CM and ICD-10-CM diagnostic codes were used to capture preoperative comorbidities that have the potential to be associated with pulmonary issues, as well as medical and surgical perioperative complications prior to the discharge of patients. According to NIS, a total of 29 variables indicate comorbidities, including AIDS, alcohol abuse, congestive heart failure, chronic pulmonary disease, coagulopathy, rheumatoid arthritis/collagen vascular diseases, chronic blood loss anaemia, depression, deficiency anaemia, uncomplicated diabetes, diabetes with chronic complications, drug abuse, fluid and electrolyte disorders, hypertension, hypothyroidism, liver disease, lymphoma, metastatic cancer, neurological disorders, obesity, paralysis, peripheral vascular disorders, psychoses, pulmonary circulation disorders, renal failure, solid tumour without metastasis, peptic ulcer disease, valvular disease and weight loss.

Patient demographics, including age, gender and race, were evaluated. Outcome indicators, such as method of admission, LOS in the hospital, overall cost of hospitalization, comorbidities prior to surgery and in-hospital mortality, were analysed. The prevalence of pneumonia, respiratory failure and PE following TSA was the primary outcome. Secondary outcomes included variables such as LOS in the hospital, cost of admission, and inpatient mortality rate. Preoperative comorbidities that may be connected to pulmonary complications were separately collected using the ICD-9-CM and ICD-10-CM diagnosis codes.

The statistical plans were finalised before analysis was conducted. The Institutional Review Board no longer required a review of the analysis to be performed in this retrospective investigation due to the deidentification of patient files and their availability in the publicly accessible NIS surgical database. The hierarchical and cluster technique was utilized to calculate the estimates of the national incidence rate of these problems, and survey weights were applied in accordance with the NIS sample design. The yearly occurrence of pulmonary problems following TSA over the period spanning from 2010 to 2019 was assessed (Fig. 1).

Annual incidence of pulmonary complications. The annual incidence of pneumonia, respiratory failure, and pulmonary embolism after elective total shoulder arthroplasty from 2010 to 2019 is shown above. The pooled incidence of any of the aforementioned complications is shown above by the graphlabeled “Any Pulmonary Complication.” The error bars show standard error of percent

Statistical analysis was conducted using SPSS (version 22.0). All tests were deemed statistically significant at a significance level of p < 0.05. The univariate relationships between independent factors and outcome variables were examined using chi-square test for categorical variables and Wilcoxon rank test for continuous variables. Binary logistic regression analysis was performed using stepwise regression to identify the independent risk factors for pneumonia, respiratory failure and PE. This analysis incorporated all factors such as demographic information supplied by NIS, characteristics of hospitals and preoperative comorbidities (Table 1). Previous studies in the area of NIS have often used large sample sizes, and the criterion for determining statistical significance at the alpha level was set at a threshold of P ≤ 0.05.