Pulmonary function deteriorates in all patients following major abdominal surgery, with a reduction in lung volumes and the development of alveolar collapse and atelectasis in 80–90 % of patients. Atelectasis leads to hypoxemia via ventilation perfusion mismatching. Its development significantly increases the risk of PRF and PPCs and may trigger an inflammatory reaction promoting bacterial growth in the lungs and bacterial translocation into the bloodstream. The main disruptions to normal respiratory function are maximal in the first hours following surgery and generally regress after 1–2 weeks.

The use of NIV following major surgery is well established as both a prophylactic and therapeutic treatment modality. Although oxygen therapy may be effective in attenuating postoperative hypoxemia, it is only a symptomatic approach that does not reverse the underlying pathophysiological process. There is evidence to suggest that lung expansion therapy using incentive spirometry and deep-breathing exercises reduces PPCs after abdominal surgery [8]. Compared with standard positive pressure respiratory therapy, the application of continuous NIV is associated with an increase in functional residual capacity and reduced atelectasis and left ventricular afterload, with a subsequent increase in cardiac output and arterial oxygenation.

Studies comparing NIV with standard therapy have generally provided positive results supporting the use of NIV postoperatively, but they are limited by small sample sizes, with a large variation in the modality of NIV, technical implementation, and timing of application seen between studies. The small sample sizes in these studies has meant that atelectasis and PaO₂/FiO₂ (PF) ratio are commonly assessed, which may not translate into clinically relevant end points such as reintubation and mortality rates. Although reductions in rates of PPCs and reintubation have been demonstrated in the literature, there is currently limited evidence regarding the impact of NIV on mortality.

A landmark multicenter, prospective, randomized control trial (RCT) of 209 patients published in 2005 by Squadrone et al. [9] demonstrated a significant reduction in reintubation, pneumonia, and sepsis rates following early hood continuous positive airway pressure (CPAP) in hypoxemic patients after elective major abdominal surgery. ICU length of stay was lower in the CPAP group, but there was no difference in hospital length of stay or in-hospital mortality between groups. In a prospective observational study involving 72 patients with severe PRF after abdominal surgery, reintubation was avoided in 67 % of patients treated with NIV [10].

A large meta-analysis of 654 patients pooled from nine studies of NIV use following abdominal surgery demonstrated that NIV use was associated with a significantly lower rate of PPCs, including atelectasis, when compared with standard medical therapy [11]. The pooled estimate of two studies using intubation as an endpoint showed a beneficial effect of CPAP (risk reduction 0.85; 95 % confidence interval (CI) 0.34–0.97). Two of the studies included in the analysis assessed the effect of postoperative CPAP on mortality following abdominal surgery; however, the number of deaths was too small to allow meaningful analysis. The studies included in this meta-analysis displayed marked heterogeneity, differing in both application and duration of CPAP use, and only included preoperatively healthy patients in five studies and may therefore underestimate the beneficial effects of NIV.

CPAP may be of particular benefit for patients who cannot participate with incentive spirometry or deep-breathing exercises. In patients with obstructive airways disease, CPAP decreases work of breathing by counterbalancing the inspiratory threshold load imposed by intrinsic positive-end expiratory pressure (PEEP). Noninvasive positive pressure ventilation (NPPV) may be considered in patients in whom hypercarbia coexists with hypoxemia, when there is a history of chronic obstructive pulmonary disease, or in patients who are experiencing an increased respiratory workload.

The optimum amount of PEEP and duration of NIV, particularly as a prophylactic treatment, remains controversial, with a lack of supporting evidence and trials in which a benefit for NIV has not been evident [7, 12]. Some authors have suggested that immediate application of NIV post extubation may be more beneficial in recruiting alveoli than delayed or intermittent NIV, although current practice varies widely and is often dictated by local preferences and protocols or the need to balance continued alveolar recruitment with patient comfort, nursing availability, and workload.

Upper gastrointestinal (GI) surgery in which a surgical anastomosis has been formed has historically been considered a contraindication to NIV because of the theoretical risk of NIV causing increased gastric luminal pressures and subsequent disruption of the surgical anastomosis. In a prospective cohort study, 1,067 patients undergoing a gastrojejunostomy as part of a gastric bypass procedure were assessed [13], and no correlation was found between anastomotic leaks and use of CPAP. In a retrospective review involving 91 patients receiving CPAP after laparoscopic Roux-en-Y gastric bypass surgery, the incidence of anastomotic leakage was zero [14]. Additionally, transmural gastric pressures have not been demonstrated to increase following the application of CPAP after laparoscopic Roux-en-Y gastric bypass [15]. Despite the lack of RCT data in this area, it would appear that NIV usage is not associated with an increased risk of anastomotic failure and is safe to use following upper GI surgery, although caution should be exercised to avoid high airway pressures (>25 cmH₂O). It would also seem prudent in NPPV use to limit the maximum Pressure support ventilation (PSV) level to 8 cmH₂O to prevent generation of excessively high pressures during inspiration.