Complications of Non-invasive Ventilation Techniques
Complications of Non-invasive Ventilation Techniques
Of 2823 reports on NIV initially identified in the literature, 1967 records were excluded because they were not relevant to the review; a further 702 reports were excluded because they were short (i.e. abstract or letter), or not reporting on complication, or reporting on paediatric subjects. One hundred and fifty-four articles were selected and used in the qualitative review. Sixty-two RCTs including a total of 5870 patients were included for subsequent meta-analyses (Table 1). A schematic of our study selection protocol is presented in Figure 1.
Table 2, Table 3 and Table 4 detail the relative incidences of diverse NIV failure causes determined from our literature search and of major and minor complications related to NIV. Suggestions for clinical interventions to prevent and deal with complications are included as well.
After assessing the RCTs (NIV vs standard medical care), the overall NIV failure occurred in 16.3% (360/2198) of patients and failure for all causes had a small but significant RR of 0.88 (95% CI: 0.85–0.91; P<0.0001, not shown) (Table 2). Our meta-analysis categorized NIV failure causes according to underlying disease states (Fig. 2). Although NIV was associated with markedly lower mean failure ORs vs standard therapy in COPD (RR 0.71, CI: 0.71–0.87, P<0.0001), hypoxic ARF (RR 0.86, CI: 0.79–0.93, P=0.0004), hypoxic–hypercapnic ARF (RR 0.84, CI: 0.75–0.94, P=0.0025), and postoperative ARF (RR 0.92, CI: 0.88–0.96, P= 0.0009), the statistical significance of these differences was not maintained after completion of and meta-analysis (Fig. 2).
(Enlarge Image)
Figure 2.
Failure rates with NIV vs standard medical therapy in ARF. The figure shows the meta-analysis for (top to bottom) COPD, ACPO, hypoxic ARF, hypoxic–hypercapnic ARF, postoperative ARF, post-extubation ARF, and weaning.
Overall NIV-associated pneumonia occurred in 5.7% (67/1172) of patients (not shown). Our meta-analysis of pneumonia incidence in NIV vs standard medical care showed no clear statistical association between NIV treatment and pneumonia in diverse medical conditions (Fig. 3). Nonetheless, pneumonia incidence comparing NIV vs standard medical care for all causes had a small but significant RR of 0.92 (95% CI: 0.89–0.94; P<0.0001, not shown). In particular, NIV was associated with markedly lower mean pneumonia ORs vs standard therapy for failure in hypoxic ARF (RR 0.89, CI: 0.82–0.95, P=0.0025) and postoperative ARF (RR 0.94, CI: 0.90–0.98, P=0.011), which may display significance in the future when additional studies are available for meta-analysis. In contrast, when NIV was compared with tracheal intubation for ventilator support weaning, NIV showed a significant risk reduction in pneumonia incidence (RR 0.79; 95% CI 0.71–0.88, P<0.0001), by meta-analysis of five relevant RCTs (Fig. 3). This suggests that NIV may be superior to typical approaches for weaning patients off ventilator support.
(Enlarge Image)
Figure 3.
Pneumonia incidence with NIV vs standard medical therapy in ARF. The figure shows the meta-analysis for (top to bottom) COPD, ACPO, hypoxic ARF, hypoxic–hypercapnic ARF, postoperative ARF, post-extubation ARF, and weaning.
All other complications ranged from 0% to 100% (Table 3). The literature did not provide sufficient data to conduct a meta-analysis on haemodynamic complications during NIV (Table 2).
Compared with standard medical care, NIV had a small but significant RR for intolerance (RR 0.91; 95% CI: 0.88–0.93; P < 0.0001), nasal lesions (RR 0.87; 95% CI: 0.84–0.90; P < 0.0001), nasal/oral dryness/congestion (RR 0.93; 95% CI: 0.89–0.97; P=0.0025), and gastric insufflation (RR 0.96; 95% CI: 0.94–0.98; P=0.0008).
Results
Paper Selection
Of 2823 reports on NIV initially identified in the literature, 1967 records were excluded because they were not relevant to the review; a further 702 reports were excluded because they were short (i.e. abstract or letter), or not reporting on complication, or reporting on paediatric subjects. One hundred and fifty-four articles were selected and used in the qualitative review. Sixty-two RCTs including a total of 5870 patients were included for subsequent meta-analyses (Table 1). A schematic of our study selection protocol is presented in Figure 1.
Table 2, Table 3 and Table 4 detail the relative incidences of diverse NIV failure causes determined from our literature search and of major and minor complications related to NIV. Suggestions for clinical interventions to prevent and deal with complications are included as well.
NIV Failure
After assessing the RCTs (NIV vs standard medical care), the overall NIV failure occurred in 16.3% (360/2198) of patients and failure for all causes had a small but significant RR of 0.88 (95% CI: 0.85–0.91; P<0.0001, not shown) (Table 2). Our meta-analysis categorized NIV failure causes according to underlying disease states (Fig. 2). Although NIV was associated with markedly lower mean failure ORs vs standard therapy in COPD (RR 0.71, CI: 0.71–0.87, P<0.0001), hypoxic ARF (RR 0.86, CI: 0.79–0.93, P=0.0004), hypoxic–hypercapnic ARF (RR 0.84, CI: 0.75–0.94, P=0.0025), and postoperative ARF (RR 0.92, CI: 0.88–0.96, P= 0.0009), the statistical significance of these differences was not maintained after completion of and meta-analysis (Fig. 2).
(Enlarge Image)
Figure 2.
Failure rates with NIV vs standard medical therapy in ARF. The figure shows the meta-analysis for (top to bottom) COPD, ACPO, hypoxic ARF, hypoxic–hypercapnic ARF, postoperative ARF, post-extubation ARF, and weaning.
Pneumonia
Overall NIV-associated pneumonia occurred in 5.7% (67/1172) of patients (not shown). Our meta-analysis of pneumonia incidence in NIV vs standard medical care showed no clear statistical association between NIV treatment and pneumonia in diverse medical conditions (Fig. 3). Nonetheless, pneumonia incidence comparing NIV vs standard medical care for all causes had a small but significant RR of 0.92 (95% CI: 0.89–0.94; P<0.0001, not shown). In particular, NIV was associated with markedly lower mean pneumonia ORs vs standard therapy for failure in hypoxic ARF (RR 0.89, CI: 0.82–0.95, P=0.0025) and postoperative ARF (RR 0.94, CI: 0.90–0.98, P=0.011), which may display significance in the future when additional studies are available for meta-analysis. In contrast, when NIV was compared with tracheal intubation for ventilator support weaning, NIV showed a significant risk reduction in pneumonia incidence (RR 0.79; 95% CI 0.71–0.88, P<0.0001), by meta-analysis of five relevant RCTs (Fig. 3). This suggests that NIV may be superior to typical approaches for weaning patients off ventilator support.
(Enlarge Image)
Figure 3.
Pneumonia incidence with NIV vs standard medical therapy in ARF. The figure shows the meta-analysis for (top to bottom) COPD, ACPO, hypoxic ARF, hypoxic–hypercapnic ARF, postoperative ARF, post-extubation ARF, and weaning.
Other Complications
All other complications ranged from 0% to 100% (Table 3). The literature did not provide sufficient data to conduct a meta-analysis on haemodynamic complications during NIV (Table 2).
Compared with standard medical care, NIV had a small but significant RR for intolerance (RR 0.91; 95% CI: 0.88–0.93; P < 0.0001), nasal lesions (RR 0.87; 95% CI: 0.84–0.90; P < 0.0001), nasal/oral dryness/congestion (RR 0.93; 95% CI: 0.89–0.97; P=0.0025), and gastric insufflation (RR 0.96; 95% CI: 0.94–0.98; P=0.0008).
Source...