The Incidence and Risk Factors of Postoperative Pulmonary Complications in Restrictive Pulmonary Disorders

Background The incidence of postoperative pulmonary complications (POPCs) in restrictive pulmonary disorders remains indefinite especially in adults. Therefore, this study is structured to evaluate the incidence and risk factors of POPCs in restrictive pulmonary disorders. Methods 2177 preoperative consultations have been prospectively evaluated from May 2015 to May 2016 in Cukurova University, Department of Chest Diseases. 60 of them (2.8%) met restrictive pulmonary function tests (PFTs) criteria and all of them were enrolled in the study. Each participant was evaluated at the 7th day and has been followed-up until 30th day after surgery. Clinical, surgical, PFT and arterial blood gas analysis parameters were evaluated to analyse risk factors of POPCs.

mortality particularly within the first postoperative week. The reported incidence ranges from 5 to 80 percent, depending on the patient population, the type of surgery, the presence of risk factors and the definitions used. (1) It has been estimated that more than 1 million POPCs occur annually in the United States (US), within 46,200 related deaths and 4,8 million additional hospitalization days (2).
Owing to the reason that POPCs contribute to significant mortality and morbidity, current research focused on evaluating the potential risk factors. Several reports showed numerous modifiable and non modifiable risk factors to date. The well known risk factors are age, male gender, smoking, American Society of Anaesthesiologists (ASA) physical status, acute respiratory infection, comorbidities, type of surgery, intraoperative ventilation strategies and several laboratory testing including pulmonary function tests (PFTs). Within the light of these factors, many risk prediction models have been developed (ARİSCAT, AZOULLAY, PERİSCOPE).
Actually, investigating PFTs at the preoperative period is not novel (3,4). A number of previous reports focused on obstructive PFTs as a risk factor of POPCs. Although surgical and anesthesiologic techniques improved, the risk of POPCs in obstructive patients (mainly asthma and chronic obstructive pulmonary disease (COPD)) is approximately 9 % (5). The American College of Chest Physicians (ACCP) guidelines recommends preoperative PFTs with a previous diagnosis of COPD or asthma (6). However, according to best of our knowledge, there is no previous report which specifically indicates the incidence and risk factors of POPCs in restrictive pulmonary disorders. The aim of this study was to evaluate the frequency and the risk factors of POPCs in patients with restrictive pulmonary impairment.

Study Population& Design
This is a prospective descriptive study. We enrolled the adult patients that have been consulted preoperatively to our department in case of any of respiratory symptoms and/or PFT abnormality and/or abnormal chest X-ray (CXR) and/or known respiratory disorders between 15 May 2015-15 May 2016. During the enrolment period 72,922 surgeries were performed in Cukurova University Balcalı Hospital (Adana, Turkey). 45,571 of the operations were performed under local anaesthesia, so they were excluded. From the remaining 2177 preoperative consultations, 60 of them (2.8 %) met restrictive PFTs criteria and all of them were enrolled in the study after assignment of the written informed consent. The study flow chart was presented in Figure 1. The institutional ethics committee approved the study (2015/42) and written informed consents were obtained from all of the participants.

Study Protocol
At the first visit the detailed clinical examination and radiographic evaluation and if required, further diagnostic tests were performed to find out the reason of restrictive PFTs. At the second visit before the operational procedure, the following data were collected; gender, age, known comorbidities and respiratory disorders, prior medications, smoking history, detailed respiratory symptom history, physical examination and ASA physical status, Cardiopulmonary Risk Index and PFT results were recorded. All patients with restrictive PFTs were further evaluated and the reason of restriction was noted.
Inhaled bronchodilators, antibiotics and appropriate treatments implanted as needed. The third visit was performed within the first 24 hours of postoperational period, and the duration of operation, the site of surgery, the type of surgery (laparoscopic vs. open), vital findings including preoperative oxygen saturation and the arterial blood gas analysis were recorded. The patient was evaluated again before discharge and later on the 7 th and 30 th postoperative day for the presence of any POPCs. All follow up was performed by the same investigator. Standardized definitions explained below were used for the diagnosis of any POPCs.

PFTs
PFTs were performed by using a calibrated Sensor Medics V-Max 20 Spirometer. None of the patients were receiving oral or inhaled short acting beta 2 agonists 8h before testing.
Baseline forced expiratory volume in first second (FEV 1 ) and forced vital capacity (FVC) was measured 3 times and the best of three measurements was recorded. Total lung capacity was measured using the helium dilution technique (Jaeger MS-PFT Analyser Unit).
The transfer factor of the lung for carbon monoxide (T LCO ) was measured using the single breath method. The results were presented as the percentages of predicted.

Restrictive PFTs
Restrictive pattern was defined as a reduced FEV 1 and/or FVC with a normal or increased FEV 1 /FVC ratio along with the decline of lung volumes and diffusing capacity (7).

Postoperative Pulmonary Complications
European Perioperative Clinical Outcome definitions have been used in the diagnosis of POPCs (8).
Early POPCs are defined as the POPCs observed within the first 7 day after surgery and late POPCs are defined as the POPCs observed within the first 30 days after surgery.

End Points
The main end point of this study was the incidence of POPCs in patients with restrictive PFTs within the first 7 th and 30 th postoperative days. Secondary outcome was evaluating the potential risk factors of POPCs in restrictive disorders. Additional clinical outcomes including length of hospital stay and mortality was also analyzed.

Statistical Analysis
All analyses were performed using SPSS 18.0 statistical software package. Categorical variables were expressed as numbers and percentages, whereas continuous variables were summarized as mean and standard deviation and as median and minimum-maximum where appropriate. Chi-square test was used to compare categorical variables between the groups. The normality of distribution for continuous variables was confirmed with the Kolmogorov-Smirnov test. For comparison of continuous variables between two groups, the Student's t-test was used. For comparison of more than two groups, One-way ANOVA was used. To evaluate the correlations between measurements, Pearson Correlation Coefficient was used. The statistical level of significance for all tests was considered to be 0.05.

Results
2117 patients were evaluated preoperatively and 60 of them (2.8%) had restrictive pattern. The basic clinical and demographic characteristics of the study group are presented in table 1. The half of surgeries was laparoscopic. 36 (60%) of surgeries was thoracic, 10(16.7%) was upper abdominal, 3 (5%) was lower abdominal and 11 (18.3%) was other surgeries. Mean surgery duration was 1.7 ±1.4 hours. All operation were under general anaesthesia, each patient have had pain relief. The mean FVC % predicted and T LCO % predicted were 60.8 ±10.6 and 56.9 ±14.0 respectively. Half of the patients have had severe restrictive PFTs.
The incidences of early and late POPCs were 10% and 11.7% respectively. The aetiologies of POPCs are listed in table 2. All patients were alive at the end of the follow-up period (30 days).
When the patients with and without POPCs compared, the demographic, clinical characteristics, smoking history, PFTs and the preoperative risk indexes were similar.
However, the preoperative arterial partial oxygen pressure (PaO 2 ) were lower in patients with POPCs (p<0.05)(

Discussion
For the first time, this study reported the incidence of early and late POPCs as 10% and 11.7 % in restrictive pulmonary disorders.
We indicated that a cut-off value ≤68mmHg for preoperative PaO 2 is an independent risk factor of POPCs. Surgery site, duration of surgery, age, previous or current smoking, preoperative PFT results, physical status, cardiopulmonary risk points, abnormal chest radiography were not significantly related with an increased risk in our study.
The frequency of POPCs in the current literature ranges between 0.2-42% independent from the PFTs (9,10,4). In an early descriptive report, the incidence was 35% following upper abdominal surgery (11). In another retrospective observational study which included 898 patients who underwent laparoscopic gastric/colorectal surgery; it was 13% (12). Another large multicenter prospective study which enrolled non cardiothoracic surgery with high risk patients reported the frequency of POPCs as 33.4% and deemed in increased mortality and longer hospital stay (13). The possible reasons of this wide range may be the heterogeneity of populations included, the different definitions used and the type and site of surgeries. As well, with the improvement in anesthetical applications, intraoperative ventilation practices and surgical procedures, the frequency of POPCs decreased in years gradually (9,10). Thus, a recent report with thoracic and upper abdominal surgeries which are well known risk factors for POPCs, 11.5 % of 314 patients who had thoracic or abdominal resection (half of them were laparoscopic) developed POPCs (14). However, none of these reports specifically evaluated PFTs.
Evaluation of pulmonary functions tests in predicting POPCs is not novel. In a prospective cohort study with COPD patients the incidence of POPCs was 24.4% and the severity of airflow limitation was strongly associated with increased POPCs risk in COPD patients (15).
A more recent trial which focused on obstructive pulmonary disorders indicated the frequency of POPCs as 8.6% in asthma and 7.9% in COPD patients (5). Large prospective studies performed with national databases, any postoperative complication (not only pulmonary) rate was reported as 25.8%. As well, COPD was independently associated with longer hospital stay, higher morbidity and mortality rates and also with higher postoperative pneumonia, respiratory failure (16,17 Predicting the POPCs is critically important. Preoperative spirometry has been used in several reports to predict POPCs as it is rational to use a non-invasive, convenient and rapid diagnostic test. In a systematic review which included large studies, abnormalities on spirometry were found to predict postoperative pulmonary complications in only trial evaluating either of these tests (1). This small trial of patients undergoing head and neck surgery showed spirometry, especially FEV1, to be predictive of POPCs only in a univariate analysis but it did not reached statistically significance in multivariate analysis (21).
Another study of upper abdominal surgery patients showed that patients with abnormal preoperative PFT results had much more POPCs than with normal PFT result (45.2% versus 24.1% respectively)(11). ACCP guideline recommends to obtain PFTs for patients with COPD or asthma and for patients with dyspnea or exercise intolerance that remains unexplained after clinical evaluation. The guideline also refuses to use of PFTs to deny any surgery with the exception of lung resection surgery (6). However, current guideline has no recommendation on patients with restrictive PFTs. In this study, PFT results were not associated with the risk of POPCs probably due to the small study sample size.
There is conflicting data about the necessity of preoperative arterial blood gas analysis in prediction of POPCs. In a large multicenter study which included patients who underwent both general and regional anaesthesia. In addition, lower preoperative arterial oxygen saturation was showed as the strongest patient related risk factor of POPCs (22 According to our study, the most common POPCs including atelectasis, pneumonia, and pneumomediastinum were similar to the literature. Respiratory failure was developed in only 1.7% of the study group. The most common POPCs are respiratory failure in many studies (4.7%), followed by pleural effusion (3.1%), atelectasis (2.4%), pulmonary infection (2.4%), bronchospasm (0.8%), pneumothorax (0.6%), and aspiration pneumonitis (0.2%) (23,27). The most common POPCs was respiratory failure-pleural effusionpneumonia and prolonged oxygen need-atelectasis among different surgery sites regardless of PFTs (14,13). Bronchospasm, pneumonia/atelectasis and respiratory failure was the most common POPCs in asthmatics and pneumonia/atelectasis in COPD patients (5).
This study has several limitations. The main limitation is the relatively small sample size which may affect precision of the estimates and underestimation of some differences of means in terms of p value. However, we know that the incidence of restrictive PFTs is about 8-12% (28). Secondly, this study included all surgeries under general anaesthesia which may be concluded as a limitation. Finally, attending physicians were not blinded to PFT results that may have introduced information bias to the study.