Page 54 - Effects and Efficacy of (Laparoscopic) Gastrostomy Placement in Children - Josephine Franken

gastrostomy site infections and higher rates of hypergranulation and leakage.11,23,24 These differences in complication rates may be caused by a difference in placement technique, as almost all studies reported solely on PEG in children.11,23,24 Also, study design may play a role. Our cohort study, as well as many earlier studies, was performed retrospectively. This may result in missing data and therefore could influence results. All patients in our cohort study, however, were followed very closely by intensive and easily accessible postoperative consultations with the specialized stoma care nurses in our institution. This resulted in very detailed reports on all postoperative events reported in our study.
Several studies have reported on the association between LGP and the incidence of postoperative GER. However, results from these studies are contradictory and well-designed studies with adequate power are lacking. 14, 16 , 25 , 26 Our cohort study showed that LGP was not associated with an overall increase in GER incidence, as the number of patients with postoperative GER symptoms remained similar (p=0.824) to the number of patients with preoperative GER symptoms. However, assessment of GER in children undergoing LGP is challenging, especially because the vast majority of children requiring long-term enteral feeding are neurologically impaired. In these neurologically impaired patients evaluation of GER symptoms is especially difficult, since symptoms are vague and often disguised by other gastrointestinal problems.27 In addition, data on GER symptoms were collected retrospectively from patient records. In case of missing information on GER symptoms, patients were excluded from analysis.
Assessment of GER was also performed by preoperative 24-hour pH monitoring. In our study, specificity of 24-hour pH monitoring to predict postoperative GER symptoms was relatively high (76.9%), but sensitivity was very low (17.5 %). This means that when the pH metry before LGP shows pathologic reflux, the patient is likely to develop postoperative GER after LGP. However, the vast majority of patients that developed postoperative GER symptoms after LGP were not diagnosed on preoperative pH monitoring. There are missing data on results of the pH monitoring since it was performed in only 60.0% of patients. GER symptoms were reported slightly more frequently in the group of patients with a preoperative pH metry compared to the group without preoperative pH metry (57.2% vs. 45.0%). However, specificity and sensitivity of the pH metry are not dependent on the prevalence of preoperative GER symptoms.
The interpretation of 24-hour pH monitoring results is restricted by certain limitations. First, no normal values are available for patients over 16 months of age. 28,29 Secondly, pH monitoring only registers the acidic reflux episodes. Non-acidic reflux episodes, which occur more frequently in children compared to the adult population, are not registered. Therefore, because of the low sensitivity and the limitations associated with 24-hour pH
52