Obesity in pregnancy leads to greater risk of pregnancy-related complications; for example, the risk of caesarean section (C-section) delivery in obese women is estimated to be twice as high as the risk in women with a normal weight (Chu et al, 2007). The prevalence of obesity in pregnant women in the UK is increasing, with 21.3% of antenatal women classed as obese, with a body mass index (BMI) over 30kg/m2, and less than half of pregnant women having a BMI within the normal range (ie 18.5kg/m2–24.9kg/m2, with 25kg/m2–29kg/m2 being classed as ‘overweight’) (Denison et al, 2019). Postoperative wound complications are also significantly higher, with Heslehurst et al (2008) identifying that obese women were 3.5 times more likely to suffer from complications such as surgical site infection (SSI) than women with a normal BMI. Obese women are at higher risk of SSI due to decreased blood flow to adipose tissue, which leads to hypoxia at the wound bed (Ye, 2011), resulting in reduced oxidative bacterial killing, therefore increasing the risk of SSI (Allen et al, 1997). Obese women are also more likely to have comorbidities such as diabetes (Denison et al, 2019), which can also negatively impact wound healing (Leth et al, 2011). In addition to SSI, other postoperative wound complications, such as haematoma and seroma formation, and wound dehiscence, occur in obese women following C-section (Orth et al, 2016).
Closed incision negative pressure wound therapy (ciNPWT) has been introduced as a prophylactic intervention to reduce C-section wound complications (Tuuli et al, 2020). Although the exact mode of action of ciNPWT is uncertain (Gillespie et al, 2021), it is thought to aid wound healing by increasing perfusion of the wound, thereby stimulating granulation tissue, and decreasing subdermal fluid collection and tension placed on suture lines (Hyldig et al, 2019). Most studies examining the effect of ciNPWT on C-section wounds measure the incidence, or risk, of SSI as the primary outcome (Gillespie et al, 2022). A systematic review by Gillespie et al (2022) concluded that the use of ciNPWT on C-section wounds led to a statistically significant reduction in SSI rates. However, that systematic review analysed randomised controlled trials (RCTs) only, which are undertaken in controlled settings and may not always represent clinically relevant results (Scifres et al, 2009). This article will discuss the evidence from both RCTs and retrospective studies on the use of ciNPWT on C-section wounds in obese women and explore the benefits and harms it may have for patient outcomes.
A review of the current literature on the use of ciNPWT was conducted by searching the Medline online database. Key terms used for this search were ‘caesarean section’ or ‘ceserean section’, ‘negative pressure wound therapy’, ‘prophylactic negative pressure’, ‘incisional negative pressure’, ‘surgical site infection’, ‘postoperative complications’, ‘surgical wound dehiscence’ and ‘wound complication’. A total of seven RCTs and three retrospective studies examining ciNPWT use on C-section wounds in obese women were identified following this search and were included in the review. The search was completed in October 2021.
RCTs examining the use of ciNPWT for SSI
The first RCT examining the use of ciNPWT (Prevena, Acelity, San Antonio, Texas) in C-section wounds found no statistically significant difference in SSI between ciNPWT and standard dressings (P>0.05) (Ruhstaller et al, 2017). However, the sample size was relatively small (n=136), and the study was executed at a single site, reducing wider generalisability. A further multicentre RCT using a much larger sample (n=876) by Hyldig et al (2019) identified that PICO ciNPWT (Smith and Nephew, Hull, UK) reduced the risk of SSI compared with standard dressings alone (P=0.007). However, a larger multicentre RCT by Tuuli et al (2020) found no significant difference (P>0.05) in rates of SSI when using Prevena ciNPWT. This study had a larger, more ethnically diverse sample (n=1624), increasing the generalisability and external validity of the results.
These three RCTs all examined both emergency and elective C-sections. A more recent RCT by Gillespie et al (2021), undertaken across multiple centres with the largest sample size to date (n=2035), discovered a 24% reduction in risk of SSI when using PICO ciNPWT, although this did not quite reach statistical significance (P=0.06). In this study, the same surgical technique was implemented for all C-sections and participants had similar baseline characteristics, increasing internal validity. However, the study excluded women who underwent urgent or emergency C-sections. Research shows that patients undergoing emergency surgery leads to an elevated risk of SSI (van Walraven and Musselman, 2013), therefore the results may not be generalisable to women undergoing emergency C-sections.
Due to the nature of the intervention, the clinical teams in all of the above RCTs could not be blinded to which treatment was applied, which could lead to observer bias whereby the researchers could be influenced consciously or unconsciously to favour the experimental intervention (Higgins and Green, 2008). Gillespie et al (2021) and Tuuli et al (2020) tried to reduce the risk of bias by ensuring the outcome assessors were blinded when reviewing participants' wounds for signs of SSI, so that their clinical assessment was not influenced by the treatment received.
Although these studies all used an RCT design, heterogeneity in the study methods suggests the results may not be directly comparable. For example, Ruhstaller et al (2017) and Tuuli et al (2020) both used Prevena ciNPWT whereas Hyldig et al (2019) and Gillespie et al (2021) applied PICO ciNPWT, showing variation in the device used. There were differences in the negative pressure delivered by the devices, with pressures ranging from 80 mmHg to 125 mmHg. However, there is little consensus on the optimum level of negative pressure (Borgquist et al, 2011), with pressures between 75 mmHg and 125 mmHg appearing to have similar effects on wound healing (Apelqvist et al, 2017). The studies using Prevena ciNPWT showed no significant difference in rates of SSI, whereas the studies using the PICO system both found a reduction in risk of SSI, suggesting the difference in devices may influence the development of SSI, though only one of the studies examining PICO reached statistical significance (Hyldig et al, 2019). There were also differences in the standard dressings used between RCTs, with Ruhstaller et al (2017) using low-adherent dressings and Tuuli et al (2020) applying gauze and adhesive tape. Gillespie et al (2021) used a mix of standard dressings whereas Hyldig et al (2019) did not specify which standard dressing was applied, demonstrating heterogeneity between study designs.
The criteria for diagnosing SSI also varied, with two studies (Tuuli et al 2020; Gillespie et al, 2021) using the US Centers for Disease Control and Prevention (CDC) criteria (Mangram et al, 1999), whereas the other studies did not report which criteria were used (Ruhstaller et al, 2017; Hyldig et al, 2019). The definition from the latter studies stipulated that antibiotics must have been prescribed to count as SSI, whereas the CDC definition does not have this stipulation, highlighting variations in diagnostic criteria. Furthermore, although all studies followed patients up to 30 days after surgery, there were differences in the follow-up methods used. Three studies scheduled telephone call reviews at day 30 (Hyldig et al, 2019; Tuuli et al, 2020; Gillespie et al, 2021), whereas Ruhstaller et al (2017) scheduled a face-to-face visit at 4 weeks post surgery. More detailed clinical information can be yielded from a face-to-face review (Bowling and Ebrahim, 2005), with telephone calls relying on the patient accurately identifying symptoms of SSI. Therefore, the study by Ruhstaller et al (2017) may have more clinically accurate diagnoses of SSI, increasing the reliability of the results.
Retrospective evidence on the use of ciNPWT for SSI
Several retrospective studies have examined the use of ciNPWT in reducing rates of SSI in obese women following C-section. Orth et al (2016) compared ciNPWT against standard dressings (n=970) in one hospital over a 21-month period and found no significant difference in SSI (P>0.05). However, the authors did not report how SSI was diagnosed or defined; therefore, there may be inconsistencies in which wounds were included as infected. Furthermore, there was no standardisation in the dressing applied, with both gauze and silver-impregnated dressings being used, reducing internal validity. Silver dressings are antimicrobial and have been found to decrease the rate of SSI in postoperative wounds (Abboud et al, 2014). Therefore, using silver dressings in the control group may have influenced the rate of SSI.
A further study by Looby et al (2018), which examined the use of ciNPWT in women with a BMI more than 40kg/m2 (n=467) over a 7-year period, identified a statistically significant reduction in SSI rates in ciNPWT-treated wounds compared with standard dressings (P=0.04). The participants had similar baseline characteristics to those in the study by Orth et al (2016), and the researchers used the CDC criteria to define SSI, increasing both internal and external validity. However, a more recent retrospective study by Buzhardt et al (2021), which examined a much larger sample size (n=4391) over 45 months, found that ciNPWT was associated with an increased risk of SSI compared with standard dressings (P<0.001). Although there were higher numbers of participants in the ciNPWT group with a BMI more than 40kg/m2 and diabetes, which could lead to an increased risk of SSI (Leth et al, 2011), the study adjusted for co-variates and although the risk of SSI reduced, it remained statistically significant (P=0.045). This finding suggests that use of ciNPWT may lead to increased harm in C-section wounds. Further prospective studies are required to examine the safety and effectiveness of ciNPWT on SSI in C-section wounds (Buzhardt et al, 2021). Due to the retrospective nature of these studies, there was a lack of control and standardisation of the treatment applied as the studies were not carried out under rigorous experimental conditions (Mark et al, 2014), thereby reducing validity of the results. As retrospective data are collected from medical records instead of physical review of the patient, there is a risk that some wound complications may have been missed (Mark et al, 2014).
Other wound complications
Wound dehiscence has been identified as a complication following C-section in obese women due to increased tension at wound edges, leading to increased tissue pressure and hypoperfusion, which can result in cell death and tissue breakdown (Wilson and Clark, 2004). It is thought that ciNPWT can be beneficial in reducing dehiscence by decreasing lateral stresses and tension at the incision site (Wilkes et al, 2012). However, several RCTs examined wound dehiscence as a secondary outcome (Wihbey et al, 2018; Hussamy et al, 2019; Hyldig et al, 2019; Gillespie et al, 2021), and all found no significant differences in rates of dehiscence between C-section wounds treated with ciNPWT or standard dressings, suggesting ciNPWT is not beneficial in preventing dehiscence. However, there were differences in the standard dressings used, with Hyldig et al (2019) and Gillespie et al (2021) both using a mix of standard dressings in the control group. In contrast, all participants in the control group in the studies by Wihbey et al (2018) and Hussamy et al (2019) had the same standard dressing applied. This is relevant as there is variation in mode of action on wound healing between different dressings, with some dressings creating a more favourable environment for healing than others (Abdelrahman and Newton, 2011). However, the use of differing dressings may reflect what occurs in clinical practice, increasing external validity.
Haematomas and seromas have also been identified as a complication of C-section (Orth et al, 2016); ciNPWT is thought to aid the endogenous removal of fluids in the subcutaneous space, thereby preventing fluid accumulation and oedema at the incision site (Kilpadi and Cunningham, 2011). Two RCTs have examined haematoma/seroma formation as a specific secondary outcome in C-section wounds (Wihbey et al, 2018; Gillespie et al, 2021). Both studies found no significant difference in incidence of haematomas or seromas between standard dressings and ciNPWT. However, the study by Wihbey et al (2018) was halted midway due to low enrolment. The study calculated 400 participants would be required to detect a statistical significance but only 166 women were analysed. Therefore, the study was underpowered and may not have had enough participants to detect a difference between the interventions.
Another emerging use of ciNPWT is to improve the aesthetic appearance of scar tissue to increase patient satisfaction and body image (Tanaydin et al, 2018). However, there are limited studies in this area due to the feasibility of long-term follow up of patients to assess how the scar tissue develops (Zwanenburg et al, 2021). One sub-study of an RCT by Hyldig et al (2020) evaluated the cosmetic effect of ciNPWT on C-section wounds in women with a BMI over 30kg/m2 at 6 months and 12 months post surgery. Objective evaluations were completed by a plastic surgeon using the Manchester Scar scale (Beausang et al, 1998), with subjective evaluations taken from patients using the Patient Scar Assessment scale (Draaijers et al, 2004). The study found no difference in long-term cosmetic outcomes between ciNPWT and standard dressings in both the evaluation by the plastic surgeon and patient-reported outcomes. The plastic surgeon was blinded to which intervention the patient had received, to reduce observer bias and increase internal validity. However, the study used a small sample size (n=316), reducing generalisability, and more than half of participants had dropped out by the 12-month evaluation, which could lead to selection bias (Hyldig et al, 2020). Further research with a larger sample size is required to examine if ciNPWT has a beneficial effect on scar quality in C-section wounds (Hyldig et al, 2020; Zwanenburg et al, 2021).
NPWT is generally regarded as a safe therapy with very few serious adverse events reported (Vikatmaa et al, 2008). However, several studies examining ciNPWT for C-section wounds identified that ciNPWT led to an increased risk of blistering to the skin, resulting in potential harm from this therapy (Gillespie et al, 2022). In the RCT by Ruhstaller et al (2017), the ciNPWT cohort were four times more likely to have blistered skin after device removal, although this did not reach statistical significance (P>0.05). The RCT by Tuuli et al (2020) found that women had a statistically significant increased risk of an adverse skin reaction (blistering, bleeding or erythema) when using ciNPWT compared with standard dressings (P<0.001). These findings contributed to investigators terminating the study before the trial had finished, due to perceived risk of harm from ciNPWT. The RCT by Gillespie et al (2021) also found women receiving ciNPWT were at increased risk of skin blistering compared with standard dressings (P=0.03). Both studies (Tuuli et al, 2020; Gillespie et al, 2021) were multicentre RCTs with large sample sizes, increasing external validity and generalisability of the results, suggesting adverse skin reactions is a potential harm of ciNPWT on C-section wounds. Gillespie et al (2021) proposed that targeted training for clinicians in the application of ciNPWT may reduce risk of blistering. If ciNPWT is applied correctly as per manufacturer guidelines, with peri-wound skin protected using a film dressing, this should minimise adverse skin reactions, such as blistering (Vaez-Zadeh, 2011). However, Tuuli et al (2020) ensured all clinicians applying ciNPWT were trained in the management of ciNPWT and deemed competent to apply the therapy.
Although a comprehensive search of the Medline database was completed for this review of the literature, this was not undertaken in a methodical way (such as a systematic or integrative review). Therefore, relevant studies may not have been discovered and included, affecting the external validity of this review.
Obese women undergoing C-section surgery are at higher risk of postoperative wound complications such as SSI, haematoma, seroma and dehiscence than women with a normal BMI (Heslehurst et al, 2008; Orth et al, 2016). Recently, ciNPWT has been introduced as an aid to healing these wounds and preventing occurrence of wound-related complications (Tuuli et al, 2020). Although it might initially appear simple to determine if ciNPWT is advantageous for these wounds, a detailed look at the studies investigating this phenomenon (as summarised in Box 1) demonstrates it is very difficult to compare results and propose a definitive answer to this question which is supported by data.
Box 1.Issues identified with the studies reviewed in this article
- Randomised controlled trial study designs, which are undertaken in controlled settings that may not always represent clinically relevant results
- Studies with small sample sizes executed at a single site reducing wider generalisability
- Studies of only elective not emergency C-sections, which therefore may not be generalisable to women undergoing emergency C-sections
- Studies where the observers were not blinded to which patients received the therapy, which could consciously or unconsciously favour the experimental intervention
- Studies using different closed incidion negative pressure wound therapy (ciNPWT) technologies, such as Prevena or PICO, making comparison between studies difficult
- Studies using differing levels of negative pressure that is being applied by the devices
- When comparing ciNPWT with standard dressings, there are differences between studies in which standard dressings are used, demonstrating heterogeneity between study designs
- Differences between studies in the control dressings used in the control group, some with silver-impregnated dressings, which may be antimicrobial, versus non-silver-impregnated dressings
- Differences in how studies diagnose/define surgical site infections
- Difference in how 30-day post-surgery follow-ups were done in studies – telephone call interview versus in-person face-to-face visits
- Lack of control and standardisation of the treatment applied as studies were not carried out under a rigorous accepted design
- Studies that examined medical records instead of a physical review of the patient
- Studies that did not have enough people enrolled as required to detect statistical significance in the results
- Studies with a high dropout rate, which could lead to selection bias
Most studies in this field focus on reducing SSI as the primary outcome (Gillespie et al, 2022). However, several large multicentre RCTs have found conflicting evidence supporting the use of ciNPWT, with some studies finding ciNPWT reduced SSI risk (Hyldig et al, 2019; Gillespie et al, 2021), and other studies finding ciNPWT led to no significant difference in SSI compared with standard dressings (Tuuli et al, 2020). Furthermore, a recent retrospective study by Buzhardt et al (2021) discovered ciNPWT increased the risk of SSI, suggesting the use of this therapy may lead to harm.
However, the study designs have low homogeneity meaning that results from these studies may not be directly comparable (Gillespie et al, 2022). Differences in ciNPWT devices used, the criteria used to define SSI and study follow-up protocol may have affected the results gained in each study (Gillespie et al, 2022). There is also little evidence to support the use of ciNPWT when examining secondary outcome measures such as wound dehiscence, haematoma/seroma formation and long-term cosmetic appearance of scars (Wihbey et al, 2018; Hyldig et al, 2020; Gillespie et al, 2021).
However, variation in study methods such as differences in standard control dressings used, and low participant enrolment, reflects low internal and external validity in these studies. Several studies have also raised the concern of potential harm in the use of ciNPWT, with two RCTs identifying that the application of ciNPWT led to a statistically significant increased risk of skin blistering compared with standard dressings (Tuuli et al, 2020; Gillespie et al, 2021). Further research into the use of ciNPWT to reduce risk of SSI is warranted, to examine if this potential benefit outweighs the harm of increased risk of skin blistering from the device (Buzhardt et al, 2021; Gillespie et al, 2021). This could be facilitated using an RCT study design that examines both emergency and elective C-sections, comparing one type of ciNPWT (set at the same negative pressure) with the same standard dressing used for all participants in the control group. The diagnostic criteria for SSI should be clearly defined, and participants should be followed up at 30 days with a face-to-face clinical review to ensure a clinically accurate diagnosis of SSI or other wound complication.
- Obesity during pregnancy is linked to increased risk of postoperative wound complications
- Closed incision negative pressure wound therapy (ciNPWT) has been introduced as a measure to prevent several of these complications
- Studies show ciNPWT may have some beneficial effect in reducing rates of SSI, however, there is little evidence to support its use for other wound complications
- Further research is needed to ascertain how effective ciNPWT is in reducing SSI before recommending this therapy for widespread use
CPD reflective questions
- Do you use closed incision negative pressure wound therapy (ciNPWT) in your area of work? Is there evidence to support its use in different types of postoperative wounds?
- Do you feel there is merit in using prophylactic cINPWT for all closed incisional wounds, despite the fact not all postoperative wounds have complications?
- What are the barriers to using NPWT in clinical practice, and how might they be overcome?