|Home | About | Journals | Submit | Contact Us | Français|
This study was aimed at identifying factors which may affect the gap length in cases of esophageal atresia with tracheoesophageal fistula (EA-TEF) and whether gap length plays any role in determining the outcome.
All consecutive cases of EA-TEF were included and different patient parameters were recorded. Plain radiographs with a nasogastric tube in the upper esophagus were taken. Patients were grouped into T1-T2; T2-T3; T3-T4; and T4 depending on the thoracic vertebral level of the arrest of the tube. During surgery, the gap length between the pouches was measured using a Vernier caliper and the patients were grouped into A, B, and C (gap length >2.1 cm; >1-≤2 cm and ≤1 cm). The operative gap groups were compared with the radiography groups and the other recorded parameters.
Total numbers of cases were 69. Birth weight was found to be significantly lower in Group A (mean = 2.14 kg) as compared to Group B (mean = 2.38 kg) and Group C patients (mean = 2.49 kg) (P = 0.016). The radiographic groups compared favorably with the intraoperative gap length groups (P < 0.001). The need for postoperative ventilation (70.83% in Group A vs. 36.84% in Group C, P = 0.032) and mortality (62.5%, 26.9% and 15.8% in Group A, B, and C, respectively, P = 0.003) co-related significantly with the gap length.
Birth weight had a direct reciprocal relationship with the gap length. Radiographic assessment correlated with intraoperative gap length. Higher gap length was associated with increased need for postoperative ventilation and poor outcome.
The first successful repair of the case of esophageal atresia with tracheoesophageal fistula (EA-TEF) by Haight and Towsley was a triumph of humankind over one of the most life-threatening disease of newborns through centuries. The advances in patient care and a better understanding of the disease has made a considerable impact on the survival rates. Beginning with a poor outcome, the current survival rate is approaching almost 100% in the western world. However, the scenario is still not as favorable in developing countries such as India, where most of the babies of EA-TEF present late with aspiration pneumonitis and high prevalence of low birth weight. Although various prognostic classifications such as Waterston, Spitz and Montreal have been proposed, none is properly applicable for predicting the outcome in these situations.[2,3,4] The gap between the two esophageal pouches is considered to be an important independent risk factor in short-term as well as long-term outcomes in cases of EA-TEF. There is a paucity of data regarding the possible factors which may affect the gap length in EA-TEF patients. The effect of prematurity, low birth weight, and other associated congenital anomalies may have an influence on gap length in EA-TEF patients; however, they have not been studied earlier.
Preoperative assessment of gap length helps in surgeon's preparedness for the procedure and prognostication of the disease. In this study, we assessed the influence of various factors that may affect the gap length in cases of EA-TEF and the impact of gap length on the outcome.
This was a prospective observational study conducted in a tertiary care children's hospital including all cases of gross type C EA-TEF from August 2014 to January 2016. The cases of pure EA were excluded in this study. Parameters recorded were the patient age at surgery, sex, gestational age, birth weight, and parental age, the order of birth, associated anomalies and antenatal finding of polyhydraminos. Plain radiograph including chest and abdomen with nasogastric tube 8 Fr in the upper esophagus was taken for all patients to confirm the diagnosis and to look for any apparent anomalies. The thoracic vertebral level of the arrest of the distal end of the nasogastric tube was noted. This corresponds to the lower end of the upper esophageal pouch. The level of clavicle depicting the first thoracic vertebrae, a presumptive assessment was carried out at the level of upper pouch in relation to the thoracic vertebra. The radiographic findings were grouped into four groups as T1-T2; T2-T3; T3-T4; and T4 depending on the thoracic vertebral level of arrest of the lower end of the nasogastric tube. All these cases underwent right posterolateral thoracotomy and the gap length between the two esophageal pouches was measured intraoperatively in centimeters using a Vernier caliper before the fistula ligation and upper pouch mobilization. The patients were divided into three groups according to the measured gap length as Group A gap length >2.1 cm (long), Group B gap length >1-≤2 cm (intermediate), and Group C gap length ≤1 cm or less (short). The influence of various parameters on the gap length in the three groups was analyzed. The measured gap length at surgery was subsequently compared with the previously recorded radiography groups. The need for postoperative ventilation was compared between the three groups. The final outcome was recorded as the proportion of survivors in the three gap length groups at the time of discharge from the hospital.
The variables were summarized using frequency tables and mean ± standard deviation supplemented by appropriate graphs. The association of qualitative variables with gap length was assessed using Chi-square test. The correlation of quantitative variables with gap length was evaluated using ANOVA and Bonferroni test. The value of P < 0.05 was considered statistically significant. SPSS version 15.0 software (IBM, NY, USA) was used for statistical analysis.
A total of 69 consecutive cases of gross type C EA-TEF were included in this study. There were 48 males and 21 females. Gestational age ranged from 32 to 41 weeks (mean = 36 weeks). Birth weight ranged from 1300 g to 3200 g (mean = 2300 g). Primary repair was done in 61 cases and diversion by cervical esophagostomy and feeding gastrostomy was done in eight cases where the primary repair was not feasible. The distribution of patients in the three gap length groups was Group A 24 (35%), Group B 26 (38%), and Group C 19 (27%), respectively, as shown in Table 1.
Associated anomalies were found in 16 cases (23%) which included anorectal malformation (4), cardiac anomalies (2), ear deformity (2), duodenal atresia (2), hypospadias (1), pelvi ureteric junction obstruction (1), vertebral anomalies (1), annular pancreas (1), VACTERL (1) and congenital dislocation of hip (1). Antenatal diagnosis of polyhydramnios was present in 11 cases (15%). History of feeding before surgery was present in 21 cases (30%). The effect of parental age, order of birth, gestational age, and age of patient were found to be statistically insignificant in the three gap length groups [Table 1]. Birth weight was found to be low in Group A (mean = 2.14 kg) as compared to Group B (mean = 2.38 kg) and Group C patients (mean = 2.49 kg), the difference being statistically significant [P = 0.016, Figure 1].
The radiographic assessment groups were compared with the three gap length groups based on intraoperative measurement and the correlation was found to be statistically significant [P < 0.001, Table 2]. The need for postoperative ventilation was significantly higher in Group A patients (70.83%) as compared to Group C patients (36.84%) [P = 0.032, Table 2]. The overall survival was 44/69 (62.5%) patients. The mortality was higher in Group A patients (62.5%) as compared to Group B (26.9%) and Group C patients (15.8%) which was statistically significant [P = 0.003, Table 2].
Applying Waterston prognostic classification by birth weight and lung condition, the survivals were 15/17 (88%), 24/35 (68%), and 5/17 (29%) in the three Waterston groups, respectively.
Table 3 summarizes the correlation of various observed parameters with the gap length groups.
In cases of EA-TEF, the desired outcome from a surgeon's perspective is to achieve a primary anastomosis for which the gap length plays a significant role. In some studies, the gap length has also been found to be an important predictor of outcome.[5,6,7] Preoperative assessment of the gap length helps in the surgeon's preparedness for the operation. Since there is a paucity of data in the literature regarding the possible factors determining the gap length in cases of EA we sought to assess the influence of different factors on the gap length and also the impact of gap length on the outcome.
The gap length can be conveniently measured in centimeters during surgery. Alternatively, it can be described as equivalent to the vertebral bodies. In this study, we measured it in centimeters by placing Vernier calipers during thoracotomy. Pure EA was excluded from our study because early thoracotomy is not the practice for their management. For the purpose of comparison, the patients were divided into three groups based on intraoperative measurement of gap length as A 24 (35%), B 26 (38%), and C19 (27%) [Table 1].
In our study, we found that a direct relationship exists between the birth weight and the gap length [Figure 1]. As birth weight increases the gap length decreases thus making anastomosis more feasible and tension free. Our patients with low birth weight had more gap length, and such a finding has not been reported earlier. This finding is in contrary to logic that a large size baby should have a larger anatomical gap compared to a proportionately small size baby. Perhaps, this could be explained by the higher proportional growth of the esophagus in a larger body weight baby.
Patients with long gap length had significantly higher mortality in this study [Table 2] similar to the reports from Brown and Tam. They also reported an increased incidence of all complications with increasing gap length in a series of 66 neonates and proposed that a classification based on gap length was more relevant in the modern era. Our long gap patients also had much higher requirement of postoperative ventilation similar to the findings of Beasley and Uchida et al. The importance of gap length on the final outcome in EA patients have been well stressed by Upadhyaya et al. and Mansur et al. However a recent study by Thakkar et al. refuted the claims of gap length been an important prognostic factor. This is because of the practice of routine post operative elective ventilation support (PEVS) and paralysis for premature infants and tight repairs as stated by the author. The British Association of Pediatric Surgeons Congenital Anomalies Surveillance System[11,12] has currently reported in the management of EA-TEF that 90% of infants receive PEVS for a median of 3 days.
The other studied parameters [Table 3] were not found to be significantly correlating with the gap length.
The factors affecting the survival in EA have been studied extensively over the last century. The prognosis of the patient is chiefly ascertained by two factors, i.e. the severity of esophageal disease and the general condition of the patient. The preoperative risks have been evaluated using Waterston, Spitz, and Montreal classifications.[2,3,4] In the western world with early diagnosis, intensive neonatal care and postoperative ventilation, survival rates of above 90% have been reported irrespective of birth weight and associated anomalies.[13,14] The previous classifications have thus been put under question in the current scenario. In this study, the overall survival rate was 63%, which was due to a large number of low birth weight babies and associated poor general conditions (75% Waterston Group B and C) and often nonavailability of intensive care support.
Our study also shows that a correlation exists between the preoperative radiographic assessments of gap length depending on the thoracic vertebral level of the arrest of the nasogastric tube (T1-T4) with the measurement of intraoperative gap length. A plain radiograph done routinely with a tube in the upper esophagus can provide useful information regarding the position of the upper pouch in relation to the level of thoracic vertebral body and the actual gap length [Figure 2]. The arrest of the tube at T1-T2 level corresponds to gap length Group A in 5/24 (20.8%) patients [Figure 3a] and at T2-T3 level in 18/24 (75%) patients. In gap length Group C, the arrest of the tube was noted at T3-T4 level in 17/19 (89.47%) patients [Figure [Figure3b3b and andc].c]. This correlation was found to be statistically significant (P < 0.001). Such a correlation has not been published before. This finding obviates the need for a preoperative computed tomography scan to determine the gap length and prevents unnecessary radiation exposure in a neonate as advised by some authors.[15,16,17]
Birth weight was found to have a direct reciprocal relationship with the gap length with a lower weight having the higher gap. Preoperative assessment of gap length by radiography with a feeding tube in the upper esophagus showing the thoracic vertebral level of the arrest of the tube (T1-T4) correlates with intraoperative finding of the gap. With higher gap length the need for postoperative ventilation increases and the outcome becomes poor.
There are no conflicts of interest.