The prevalence of pulmonary complications after thoracic and abdominal surgery and associated risk factors in patients admitted at a government hospital in Harare, Zimbabwe-a retrospective study

It is estimated that surgical conditions account for 11% of the total global burden of disease and 25 million disability-adjusted life years (DALYs) in Africa, the region with the highest concentration of surgical DALYs (38/1000 population) (Ozgediaz and Riviello 2008; Ärnlöv 2016). According to the World Health Organization (WHO), DALY is a measure of overall disease burden, expressed as the number of years lost due to ill-health, disability or early death. Human Immunodeficiency Virus (HIV) infection complicates surgical practice by increasing the occurrence of surgery related disease conditions, presenting challenges in the surgical management of HIV infected patients and necessitating the introduction of appropriate methods to improve surgical outcomes (Odimba 2010). Furthermore, these patients also present with other comorbidities such as hypertension, diabetes, and tuberculosis which further complicate their post-operative management (Hanekom, Coetzee and Faure 2006).

Surgical and trauma patients invariably experience a sudden and profound systemic insult that often leads to decompensation of subclinical conditions and sets the stage for new and potentially life-threatening complications (Sawyer and Leon 2010). In Nigeria, the majority of post-surgical complications were related either to surgery or anesthetic care, leading to a very high mortality rate in surgical units (Okafor 2009). Post-operative complications, involving the respiratory tract remain the most frequent cause of morbidity contributing to significant increases in patient discomfort, increased length of hospital stay (LOS), increased use of resources and overall hospital costs as well as mortality (Brooks-Brunn 1995; Arozullah et al., 2001; Hulzebos et al., 2003; Reeve 2008). A post-operative pulmonary complication (PPC) is defined as “any pulmonary abnormality occurring during the postoperative period and resulting in clinically significant disease or dysfunction, adversely affecting the clinical course of the primary condition” (Sharafkhaneh et al. 2008) and the specific complications were defined by Jammer et al., (2015).

Patients undergoing abdominal and thoracic surgery are most prone to PPCs since the incidence of PPCs is inversely related to the distance of the surgical incision from the diaphragm (Rudra and Das 2006). Besides the type of surgery, increased number, and complexity of thoraco-abdominal surgical procedures, increasing severity of illness and age of patients undergoing surgery are all directly related to the patients’ post-surgical outcomes (Hulzebos et al., 2003) and to the incidence of PPCs. It is therefore important to understand the most common PPCs in our population so that preoperative assessment can partly focus on the risk factors for the PPCs. The main aim of this study was to determine the pulmonary complications developing after abdominal and thoracic surgery and the associated risk factors. It is envisaged that knowledge of such will help put strategies in place to minimize the occurrence of the complications as well as effectively manage them thus reducing morbidity and mortality rates.

Studies conducted in low-income countries showed that 10–15% of hospital admissions were due to diseases amenable to surgical treatment, of which diseases were more prevalent in the young population compared to high income countries (Hanekom, Coetzee and Faure 2006, Frankel et al., 2007, Semer, Sullivan and Meara 2010, Chalya et al., 2011, Pearse et al., 2012, Kodra, Shpata and Ohri 2016) and the same pattern was noted in the current study. The affected age group in African countries of between 15 and 44 years is the predominant economically active segment of the population. Previous reports indicated that in this age group traumatic injuries from road traffic crashes or assault were the second most common cause of morbidity and mortality after HIV infection, particularly in males (Hanekom, Coetzee and Faure 2006, Ntumba and Juma 2011). However, our study sample showed lower prevalence of HIV in both elective and traumatic surgery patients when compared to reports from other African countries (Hanekom, Coetzee and Faure 2006, Odimba 2010, Pandie et al., 2012).

The majority of patients had abdominal surgery mainly due to infection or the presence of tumors. This compared well with reports from other countries like India, Malawi, and Ethiopia where abdominal surgery was the most common surgery performed especially in the central hospitals (Bhat, Shinde and Chaudhari 2006, Lavy et al., 2007, Denu et al., 2015). The commonest causes for laparotomies in African countries have been listed as injuries, infections and neoplasms (Grimes et al., 2012, Tadyanemhandu and Manie 2015). In Zimbabwe, the prevalence of violent or injury related surgeries have been reported to be higher in the Southern part of the country, in the Matebeleland Region as compared to the Northern part (Tadyanemhandu and Manie 2015). The current study was done in the Northern part of the country, hence the small number of injury related surgeries.

Post-operative pulmonary complications were very high indicating the burden of morbidity experienced by patients undergoing thoracic and abdominal surgery in a low income country hospital. The final outcome of critically ill surgical and trauma patients depends, to a large extent, on events that take place after the original injury, surgical emergency or elective procedure (Sawyer and Leon 2010). A significant proportion of patients in our study developed various PPCs which further impacted on the duration of stay in either critical care or general hospital stay potentially worsening final outcomes.

The prevalence of PPCs in our study was found to be 42.4%. The prevalence rate for development of PPCs in our study falls in the range of 5–80% which was reported for non-thoracic surgeries by Fisher, Majumdar and McAlister (2002). The reason for this wide variation was attributed to the set-up, pre-operative and intraoperative risk factors (Rock and Rich 2003), use of different diagnostic criteria or definitions for PPCs, the type of surgery category and the type of patients recruited (Smetana 2009). The reason why our prevalence rate was relatively high might be due to the surgical specialties under consideration, since it has been reported that the prevalence of PPCs is high in abdominal and cardiothoracic surgery patients. Nosocomial pneumonia was the most common PPC identified in our study. This was similar to other studies which highlighted that pneumonia was the most common PPC as it ranked third after urinary tract infection and wound infection (Inoue et al., 2013; Denu et al. 2015). The other complications included atelectasis, respiratory failure, and pleural effusion which were also reported in other studies (Jensen and Yang 2007, Okafor 2009).

To reduce the incidence or prevalence of PPCs, there must be an understanding of patient conditions that increase the risk of developing PPCs and of effective interventions available to reduce the impact of pre-existing patient’s conditions on the subsequent development of PPCs (Rudra and Das 2006, Denu et al., 2015, Kodra, Shpata and Ohri 2016). The risk factors for development of PPCs from our study were found to be incision type, presence of comorbidities, duration of surgery, length of critical care stay, and hospital stay. These findings were also reported in other studies which looked at predictors of PPCs in patients (Rudra and Das 2006; Jensen and Yang 2007; Denu et al. 2015, Kodra, Shpata and Ohri 2016). Gender, age over 65 years and category of surgery were not found to be associated with development of PPCs in our study contrary to other studies (Jensen and Yang 2007, Denu et al., 2015, Kodra, Shpata and Ohri 2016). This may be due to a small number of patients over the age of 60 in our study as majority was below the age of 45. However, there is evidence that advanced age is an important predictor of postoperative pulmonary complications, even after adjustment for comorbid conditions (Qaseem et al. 2006). Moreover, following any type of surgery, patients are reluctant to move because of fear of damaging the incision site and the pain associated with any movement. This predisposes the patients to the complications of immobility. Development of PPCs will further result in prolongation of mechanical ventilation, CCU stay or hospital stay, and this was evident in our study during comparisons between the group which developed PPCs and the one which did not. It is assumed that this resulted in substantial economic burden on the hospital resources. Surgical patients from 28 European countries have been reported to have a median critical care stay of 1.2 days (IQR = 0.9–3.6) and a median duration of hospital stay of 3.0 days (IQR = 1.0–70), far lower than what is reported in most African countries (Hanekom et al., 2006, Chalya et al. 2011, Tadyanemhandu and Manie 2015). The mortality rate for our study was comparable to the 12.3% reported in South Africa (Hanekom et al., 2006) but lower than 32.7% reported in Tanzania. A plausible reason for the high mortality in Tanzania was the fact that the study was focused on trauma admissions only and not any other cause. Furthermore, high income countries reported very low mortality rates as exemplified by the 4% obtained in the study of 28 European nations (Pearse et al., 2012). After controlling for other variables in our cohort, the only risk factors associated with development of PPCs in our study were duration of surgery and length of hospital stay.

The most common comorbidities present in our study were HIV and hypertension. These comorbidities have been reported in South Africa (Hanekom, Coetzee and Faure 2006, Pandie et al., 2012). The greatest burden of disease in Africa comes from infectious diseases, with HIV and tuberculosis (TB) dominating the health landscape but now there is an increase in non-communicable diseases, with hypertension and diabetes dominating (Pandie et al., 2012). There is sufficient evidence to support that exercise capacity, diabetes, and HIV infection are independent risk factors for postoperative pulmonary complications (Hulzebos et al., 2003; Rudra and Das 2006). These comorbidities predispose patients to post-operative pulmonary complications as they result in patients having a poor baseline pulmonary function.

Results of our study show that none of the patients received pre-operative physiotherapy and the majority only received post-operative physiotherapy after a complication had developed. Similar findings were obtained in Australia and New Zealand where very few physiotherapists reported involvement in pre-operative physiotherapy despite the significant benefits highlighted in the literature (Reeve 2008). The short period between diagnosis and surgery has often been regarded as an inadequate period of time to sufficiently impact upon exercise capacity in order to improve the outcomes of patients post-surgery (Reeve 2008). Reductions in post-operative pulmonary complications will require development and consensus acceptance of diagnostic criteria and creation of valid standard of measurement of the complications (Romig and Dorman 2011).

Post-operative pulmonary complications were common and were associated with increased morbidity and mortality and adversely affected clinical outcomes of patients. The presence of the comorbidities which included HIV, hypertension, and diabetes further complicated the management of the patients. The risk factors associated with development of PPCs in our study were duration of surgery and length of hospital stay. Strategies to reduce the rate of the complications have to be implemented by the health practitioners as a team in order to reduce the burden of disease on the patients as well as on hospital management.