The following inclusion criteria were applied: (1) a pathologically confirmed diagnosis of stage IA or IB gastric cancer[7]; (2) first-time gastrectomy status; (3) age between ≥ 20 and ≤ 75 years; (4) no history of chemotherapy; (5) no evidence of recurrence or distant metastasis; (6) gastrectomy conducted ≥ 1 year prior to study enrollment; (7) performance status (PS) ≤ 1 on the Eastern Cooperative Oncology Group (ECOG) scale; (8) full ability to understand and respond to the questionnaire; (9) no history of other diseases or surgeries that might influence responses to the questionnaire; (10) absence of organ failure or mental illness; and (11) provision of written informed consent. The exclusion criteria included patients who had an additional malignancy and patients who underwent the concomitant resection of other organs (except for cholecystectomy or splenectomy).

PGSAS-45 is a newly developed, multidimensional QOL questionnaire (QLQ) consists of a total of 45 questions, with 8 items from the Short-Form Health Survey (SF-8)[8], 15 items from the Gastrointestinal Symptom Rating Scale (GSRS)[9], and 22 clinically important items selected by the JPSWP⁵. These important items consists of newly selected 10 PGSAS specific items for symptoms (items 24-33), eight questionnaire items pertain to dietary intake (items 34-41), one item for social activity (item 42) and three items about level of satisfaction with daily life (items 43-45). For the 23 symptom items, a seven-grade (1-7) Likert scale was used. A five-grade (1-5) Likert scale was used for all the other items, except for items 1, 4, 29, 32, and 34-37. For items 1-8, 34, 35, and 38-40, higher scores indicate better conditions. For items 9-28, 30, 31, 33, and 41-45, higher scores indicate worse conditions.

This study utilized continuous sampling from a central registration system for participant enrollment. The questionnaire was distributed to all eligible patients as they presented at the participating clinics. The patients were instructed to mail the completed questionnaires directly to the data center. All the QOL data from the questionnaires were matched with individual patient data collected via case report forms (CRFs) sent from the physicians in charge of the patients and stored in a database.

This study was registered with the University Hospital Medical Information Network’s Clinical Trials Registry (UMIN-CTR; registration number 000002116). The study was approved by local ethics committees at each institution. Written informed consent was obtained from all the enrolled patients.

PGSAS-45 questionnaires were distributed to 2922 patients between July 2009 and December 2010. Among the 2520 (86%) patients who returned completed questionnaires, 152 were determined to be ineligible because of age (older than 75 years, n = 90), postoperative period > 1 year (n = 29), resection of other organs (n = 8), or other factors (n = 25). The data and responses from the remaining 2368 patients (81%) were eligible for subsequent analyses. In the current study, data from 868 patients who underwent either total gastrectomy (TGRY, n = 393) or distal gastrectomy (DGRY, n = 475), all with Roux-en-Y reconstruction, were retrieved from the database and analyzed. Other data from 1500 patients who underwent distal gastrectomy with Billroth-I reconstruction (DGBI, n = 909), pylorus preserving gastrectomy (PPG, n = 313, proximal gastrectomy (PG, n = 193) or local resection (LR, n = 85) were excluded for this study.

Based on the data from the completed PGSAS-45 questionnaires, the outcome measures were refined by consolidation and selection[5]. The 23 symptom items were consolidated into the 7 symptom subscales (SS) listed in Table 1. The main outcome measures for the assessment data included several subscales such as the total symptom score, quality of ingestion, level of dissatisfaction for daily life, a physical component summary (PCS) based on items derived from the SF-8, and a mental component summary (MCS), also based on SF-8 items. Each SS score, except the PCS and MCS, was calculated as the mean of the composed items, and the total symptom score was calculated as the mean of seven symptom SSs. In addition, the following parameters were selected as the main outcome measures: body weight changes, amount of food ingested per meal, necessity for additional meals, ability for working, dissatisfaction with symptoms, dissatisfaction at the meals, and dissatisfaction at working.

In comparing patient QOLs after TGRY and DGRY, statistical methods included the t-test and χ² test. All outcome measures that exhibited significant difference in univariate analysis were further analyzed using multiple regression analysis to eliminate confounding factors by adding time interval from surgery, age, gender, surgical approach, and celiac branch preservation to the type of gastrectomy. P < 0.05 was considered statistically significant. In the case of P < 0.1 by univariate analysis, Cohen’s d was calculated. In the case of that P value of standardized regression coefficient (β) in multiple regression analysis was < 0.1, the β value has shown in the table. Cohen’s d, β, and R² (coefficient of determination) measure effect sizes. Interpretation of effect sizes were 0.2 < small, 0.5 < medium, and 0.8 < large in Cohen’s d; 0.1 < small, 0.3 < medium, and 0.5 < large in β; and 0.02 < small, 0.13 < medium, and 0.26 < large in R². StatView software for Windows Ver. 5.0 (SAS Institute Inc.) was used for all statistical analyses.

The demographics of all the study participants enrolled from 52 institutions are listed in Table 2. Among the patients who underwent TGRY, the time interval from surgery until the current evaluation and the length of the Roux segment were significantly longer; age, incidence of applying an open approach, extent of lymph node dissection, and combined resection rate were significantly higher; and a posterior route of Roux-en-Y was significantly more common.

PGSAS symptom subscales: Twenty-three symptom items comprising items derived from the GSRS and original items proposed and selected by the participating gastric surgeons during the establishment of the PGSAS were consolidated into seven symptom SSs (i.e., the PGSAS symptom SSs). These SSs reflected the PGS symptom profile and revealed that the patients who underwent TGRY had significantly higher (i.e., worse) scores (Table 1 and Figure 1A). The implication of the type of gastrectomy on each PGSAS symptom SS, in terms of effect size “Cohen’s d”, is shown in Table 1. To eliminate confounding factors, a multiple regression analysis was performed by adding the time interval from surgery, age, gender, surgical approach, and celiac branch preservation as explanatory variables. Cohen’s d (by the univariate analysis) and β (by the multiple regression analysis) of the esophageal reflux SS, meal-related distress SS, indigestion SS and dumping SS were significantly declined by the extent of gastrectomy with medium to small effect size (e.g., Cohen’s d and β, Tables 1 and 3).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Rader charts of Postgastrectomy Syndrome Assessment Scale symptom subscales, life status, and dissatisfactions. A: Twenty-three symptom items consisting of Gastrointestinal Symptom Rating Scale and Postgastrectomy Syndrome Assessment Scale (PGSAS) specific items were consolidated into seven symptom subscales as PGSAS symptom subscales (SS). This radar chart demonstrated the worse conditions of postgastrectomy syndromes (PGS) in total gastrectomy (TGRY) group than that in distal gastrectomy (DGRY) group; B: The radar graph consisting of each score of items related to the life status and the dissatisfactions demonstrated the worse conditions of PGS in TGRY group than that in DGRY group.

The total symptom score, which aggregated all seven PGSAS symptom SSs, denoted the severity of overall PGS symptom and its values of Cohen’s d, β, and R² in the present study were 0.38, 0.216, and 0.059, respectively. This also indicated that the severity of overall PGS symptom after TGRY were significantly greater than those after DGRY (Tables 1 and 3).

Body weight changes: The body weight change more than one year after gastrectomy was -13.8% in the patients who underwent TGRY and -8.9% in those who underwent DGRY. The type of gastrectomy had a significant and medium implication on body weight loss as to effect size; the uni-variate analysis produced a Cohen’s d of 0.66 (Table 1), and the multiple regression analysis produced a β of 0.315 (Table 3).

Meals, social activity, and dissatisfaction: The ingested amount of food per meal, necessity for additional meals, ability for working, and dissatisfaction with symptoms at the meal and at working were important items included in the PGSAS-45. The scores on these items revealed that the patients who underwent TG experienced significantly worse living status and more dissatisfaction for their lives (Table 1 and Figure 1B).

The Cohen’s d values for the necessity for additional meals, dissatisfaction at the meal and dissatisfaction for daily life SS were of medium effect size. All the other items related to dissatisfaction had small Cohen’s d values, indicating that they were significantly influenced by the type of gastrectomy with small but clinically meaningful effect size (Tables 1 and 3). The dissatisfaction for daily life SS, which consisted of three dissatisfaction items, demonstrated a significant influence of PGS. The Cohen’s d, β, and R² values for this SS were 0.51, 0.268, and 0.083, respectively (Tables 1 and 3).

SF-8: The SF-8, a useful and simple questionnaire for evaluating generic QOL, includes PCS and MCS as SSs. For the PCS, Cohen’s d was 0.21, and β was 0.109. Thus, the PCS was significantly lower in the TGRY cases (with a small effect size), whereas the MCS did not differ between the TGRY and DGRY cases (Tables 1 and 3).

The influence of postgastrectomy syndrome (PGS) after total gastrectomy (TG) is believed to be more intense than that after distal gastrectomy (DG). However, the precise features and the degree of interference with quality of life (QOL) after TG against DG have not been clarified. It was discussed by a comparison between TG and DG in the past report, however, even if it was the same DG. Most of prior reports compared the postoperative QOL between TG with Roux-en-Y (RY) and DG mainly with Billroth-I (BI) or II (BII), in which the food or digestive juice passes through the other route. Therefore, influence due to the physiological reconstruction might be mixed in the difference of both as well as influence of TG with RY (TGRY).

Most of prior reports compared the postoperative QOL between TG with RY and DG mainly with BI or BII, in which the food or digestive juice passes through the other route. Several reports have postulated that preserving the lower esophageal sphincter or cardia prevents esophageal reflux. Moreover, esophageal reflux has been less frequently observed following DGRY than after DGBI. Thus, preserving the proximal stomach with Roux-en-Y reconstruction should help to prevent esophageal reflux.

This study was to have compared the DG with TG except influence of the reconstruction course. And we examined influence of the presence or absence of reservoir capacity by multivariable analysis including other clinical factors, and showed the statistical effect sizes. Moreover, this study was unprecedented examination with large number cases and used newly developed PGSAS-45 questionnaire evaluating patient QOL following gastrectomy. The PGSAS-45 defined a certain impairment of QOL after TGRY compared to DGRY specifically. Therefore, it is important to closely monitor patients to enable early detection and management of PGS, and to provide appropriate nutritional support. The results of the present study defined that the sole loss of remaining stomach strongly and broadly impairs the QOL in postgastrectomy patients.

The study results suggest that reservoir reconstruction using a stomach substitute should be an optional alternative to maintain better QOL after TG.

Postgastrectomy syndrome is a group of disorders and complications following gastrectomy. It includes early dumping syndrome, late dumping syndrome, bile reflux gastritis, afferent loop syndrome, efferent loop syndrome, Roux syndrome, postvagotomy diarrhea, malabsorption, anemia, osteoporosis, gastroparesis, and weight loss.

The authors have conducted a well-written study based on sound methods. The manuscript flows easily and with clarity. The case selection and choice of the important variables were appropriate and clinically relevant. The graphs and the tables are to the point and summarizing the main results of the study in an appropriate fashion. The retrospective nature limits the study, and the analysis would benefit from a propensity score matching to match the cases better and hence give stronger inference.