Important progress and development have been made in the contemporary medical methods and anesthesia technology, as a result of the amount of neonatal abdominal surgical operations increasing rapidly[15, 16, 18]. Due to their special pathophysiological and anatomical characteristics, neonates are vulnerable to AWD[12, 16]. Neonatal low immunity, improper intraoperative operation and postoperative abdominal distension as reported before[19, 20] are risk factors for neonatal AWD, which are related to the prognosis of neonatal AWD, but still cannot be used as biological indicators. Preoperative comprehensive assessment of the patient’s overall nutritional status and appropriate perioperative operation has been considered to reduce the complications. A recent study has discovered that hypoproteinemia and contaminated incision, two independent risk factors for neonatal AWD, can be considered as feasible clinical predictors easy to perform[16], the relationship between which and neonatal AWD and their inner relationship still remain to be discussed so far. In this context, this study used Cox proportional-hazards model (the Cox model) to analyze the relations of hypoproteinemia, contaminated incision and their combination with neonatal AWD in the training set and validation set separately and to compare predictive value of those for neonatal AWD respectively.
The incidence of wound dehiscence after deep infections has been reported 5 to 10 times higher compared with wound after primary healing[21]. However, we discovered that the rate of contaminated incisions was 3.79 times higher than that of clean wounds. By contrast with clean wounds (class I), the incidence of AWD with class II incision increased (4.9% vs. 1.9%), while the incidence with class III incision was significantly higher (12.9% vs. 1.9%), In another words, a significant growing rate of incision dehiscence with the increasing severity of contamination suggested that wound contamination was an exact risk factor for AWD, which may the result of rising degree of incision contamination and growing number of bacteria in the incision area enabling patients to get infection and the wound difficult to heal once infected.
Hypoproteinemia, giving rise to widely spread indicator of albumin to evaluate nutritional status is exquisitely linked to surgical and postoperative complications[22–24]. Our data indicated a significantly higher incidence of neonatal AWD with hypoproteinemia (12% vs. 3.9%) based on the full Kaplan-Meier survival analysis, demonstrating the tight relationship between hypoproteinemia and neonatal AWD, which is consistent with Kim J. et al. corroborated other studies involving detrimental healing of abdominal wounds caused by hypoproteinemia[25]. It reduces not only the synthesis of collagen and collagenase, but also the development of granulation tissue on the wound surface, leading to poor wound healing. Nevertheless, multiple system organs functions are ruined by hypoproteinemia, especially immune system, adversely affecting the influx of cells and compounds necessary for tissue repair, resulting in the slump anti-infection ability of injured skin[26–29].
In Duan et al article[12] where univariate and multivariate binary logistic regression analyses were conducted representing hypoproteinemia or contaminated incision could be applied as a clinical predictor of neonatal AWD. It is rational to presume that combined factor would perform better as substitute for potential predictive indicator. To ensure a comprehensive evaluation of influencing factors, we collected and analyzed data using univariate and multivariate Cox analysis in training and validation set. The Cox proportional-hazards model (the Cox model), as a valuable decision-support system, was used in this study due to its capacity for analysis of both hypoproteinemia and incision type relationships with neonatal AWD and because it can address whether combination of those factors is more important for predicting. The ROC curve analysis showed that AUC value of the combination of both is higher and a significant difference between the hypoproteinemia factor (P = 0.0023) by Z test, but no significantly different between combination of both and contaminated incision group (P = 0.0557). It shows that contaminated incision has a greater influence on wound healing, which is consistent with our clinical work. Combination of two factors have been found to be comparable and superior to traditional single indicator in performance and also improved the accuracy and consistence of prediction. Although the accepted perception that either hypoproteinemia or contaminated incision will provide increased opportunities for AWD, both of them existed in the result of AWD, which remained to be considered that their joint effort should be to blame. When both of hypoproteinemia and contaminated incision occurs, patients with hypoproteinemia tend to decrease the colloid penetrations and increase the wound exudation, providing a medium for the growth and reproduction of local bacteria in contaminated incision, which contributes to poor wound healing[30]. At meanwhile, hypoproteinemia threatens immune function and expands the opportunities of acquiring infection, equal to the prolongation of inflammatory period, which further promotes the consumption of albumin inside and negatively related to the accumulation of collagen in the wound and activity of fibroblasts[31–33]. To conclude, all these theories suggest that hypoproteinemia combined with incision contamination in neonates performs better in prediction for incidence of AWD, enhancing efficiency and ensuring accuracy.