Predictors of postoperative morbidity and delayed initiation of adjuvant chemotherapy after cytoreductive surgery in advanced-stage epithelial ovarian cancer

Abstract

Objective

This study aimed to identify the factors predicting postoperative morbidity after cytoreductive surgery in patients with advanced-stage epithelial ovarian cancer, to evaluate the effect of this morbidity on the time to initiation of adjuvant chemotherapy, and to examine the association between delayed chemotherapy and survival outcomes.

Methods

Patients with FIGO stage III–IV epithelial ovarian cancer who underwent surgery at a tertiary university hospital between May 2020 and December 2024 were retrospectively evaluated. Severe postoperative complications were defined as Clavien–Dindo grade IIIa or higher. Initiation of adjuvant chemotherapy later than 42 days after surgery was considered delayed. Clinical, laboratory, and perioperative variables associated with postoperative complications were analyzed using univariable and multivariable logistic regression analyses. Survival analyses were performed separately in the primary debulking surgery (PDS) and interval debulking surgery (IDS) groups.

Results

A total of 180 patients were included in the study, and 20 (11.1%) developed postoperative complications. In comparative analyses, bowel surgery (OR 3.55, 95% CI 1.32–9.34, p = .010), perioperative transfusion (OR 5.28, 95% CI 1.84–19.05, p = .004), and postoperative transfusion (OR 3.17, 95% CI 1.20–9.34, p = .025), were found to be significantly more frequent in the group that developed complications. In multivariable analysis, only perioperative transfusion was identified as an independent risk factor for postoperative complications (OR: 3.72; 95% CI: 1.17–14.25; p = .035). No significant association was found between the time to initiation of adjuvant chemotherapy and most patient- and surgery-related variables; however, chemotherapy was initiated earlier in patients who underwent IDS. In survival analyses, no significant association was observed between the timing of chemotherapy initiation and progression-free survival or overall survival in the PDS group, whereas in patients undergoing IDS, early initiation of chemotherapy was associated with significantly improved overall survival (p = .0013).

Conclusion

In advanced-stage epithelial ovarian cancer, morbidity after cytoreductive surgery appears to be more closely related to surgery-related factors. Timely initiation of adjuvant chemotherapy may be particularly important for overall survival, especially in patients undergoing IDS.

Introduction

Ovarian cancer is one of the most lethal gynecologic malignancies. According to cancer statistics published in 2022, more than 324,000 new cases were reported, and more than 206,000 deaths occurred within one year [1]. Patients with ovarian cancer usually present with nonspecific symptoms such as fatigue, bloating, and vague abdominal pain [2]. These symptoms, which make diagnosis difficult, often lead to the disease being diagnosed at an advanced stage, when it has spread beyond the pelvis [International Federation of Obstetrics and Gynaecology (FIGO) stage III and IV] [3–5]. Diagnosis at an advanced stage also results in low 5-year survival rates [6].

Cytoreductive surgery is one of the main components of treatment in newly diagnosed ovarian cancer. The aim of cytoreductive surgery is complete macroscopic resection [7, 8]. When complete resection of all visible tumor is achieved, it is an important prognostic factor for survival [8]. According to the National Comprehensive Cancer Network (NCCN) guidelines, the recommended treatment for advanced-stage ovarian cancer (FIGO stage III–IV) is primary debulking surgery (PDS) followed by chemotherapy. In patients who are not suitable candidates for surgery or who are unlikely to achieve optimal cytoreduction, neoadjuvant chemotherapy (NACT) followed by interval debulking surgery (IDS) is recommended [9, 10]. In order to obtain the full benefit from PDS, aggressive surgery is required, which may lead to higher complication rates [11]. Better survival outcomes have been reported in patients who achieve complete PDS [12]. Compared with PDS, IDS is a less demanding, less complex surgical procedure [13]. Although debates over survival outcomes are ongoing, studies have reported similar outcomes [14]. For these reasons, as stated in the NCCN guidelines, patient selection should be made carefully.

Complications may occur after cytoreductive surgery. These postoperative morbidities may cause delayed initiation of adjuvant chemotherapy [15]. Some studies have shown that delayed initiation of adjuvant treatment negatively affects prognosis [16–18]. Although there is still debate about the optimal timing of adjuvant treatment, it is generally recommended to initiate treatment within the first 6 weeks to improve prognosis [18, 19].

This study aimed to determine whether complications can be predicted during preoperative evaluation and which patient groups are at higher risk of developing complications. It also aimed to evaluate the effects of patient-related preoperative characteristics and postoperative complications on the time to initiation of adjuvant chemotherapy, as well as the possible effect of delayed chemotherapy on prognosis. Through this data, we aimed to contribute to the identification of preoperative risk factors and to the optimal timing of adjuvant chemotherapy in appropriate patients.

Materials and methods

Study design and setting

This study was approved by the Ethics Committee of Başakşehir Çam and Sakura City Hospital (Approval number: E-96317027-514.10-287954128). The study was carried out at Başakşehir Çam and Sakura City Hospital, a tertiary care center in Turkey accredited by the European Society of Gynaecological Oncology (ESGO). Patients who underwent surgery between May 2020 and December 2024 were retrospectively reviewed. The study was carried out in accordance with the Declaration of Helsinki and the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Study population

The inclusion criteria of the study were defined as women older than 18 years of age who were admitted to our hospital, had stage III or IV epithelial ovarian cancer according to the current FIGO staging system, and underwent surgery [20].

The exclusion criteria were as follows: patients with a history of malignancy, those who had previously received chemotherapy or radiotherapy, those with an Eastern Cooperative Oncology Group (ECOG) performance status ≥ 3, pregnant women, and patients with missing data.

The diagnostic and treatment processes for the patients were managed by physicians specializing in gynecologic oncologic surgery. The decision between primary debulking surgery and neoadjuvant chemotherapy followed by interval debulking surgery was made by a multidisciplinary tumor board based on clinical condition, radiologic disease extent, and the estimated likelihood of achieving complete cytoreduction. A standard platinum-based chemotherapy regimen (carboplatin and paclitaxel) was used. Neoadjuvant chemotherapy was administered in the IDS group, whereas adjuvant chemotherapy was given in both PDS and IDS groups. Patients who received NACT underwent surgery within 4–6 weeks after receiving 3 or 4 cycles of treatment. The decision to perform HIPEC was made on a case-by-case basis by a multidisciplinary tumor board based on clinical condition and disease extent. When delayed initiation of adjuvant chemotherapy was evaluated, patients who declined treatment by choice and those who received treatment at other centers and were lost to follow-up were excluded from the analysis. Patients with suboptimal cytoreduction were excluded from the analysis of delayed chemotherapy because residual disease may independently affect the postoperative treatment plan and the timing of chemotherapy.

Data and variables

Demographic data, pathology results, operative notes, the clinical course during hospitalization, and post-discharge follow-up notes were retrieved from the hospital system and recorded. Laboratory results were recorded from blood tests performed within the 7 days prior to surgery.

The first outcome evaluated in this study was postoperative complications. The Clavien–Dindo classification (CDC) was used to assess complications [21]. Severe complications were defined as Clavien–Dindo grade IIIa or higher and included life-threatening conditions occurring within the first 30 days after surgery, such as invasive radiologic intervention, reoperation, admission to the intensive care unit, single or multiple organ failure, or death. The Aletti score was used to evaluate the degree of surgical complexity [22]. According to this scoring system, based on points calculated from operative notes, a score of 3 or less indicates low surgical complexity, a score of 4–7 indicates intermediate surgical complexity, and a score of 8 or higher indicates high surgical complexity.

As a secondary outcome, delayed initiation of adjuvant chemotherapy was evaluated. Patients who could not start chemotherapy within 42 days after surgery were considered to have delayed chemotherapy. A 6-week cutoff was selected based on previous studies and its common use as a clinically meaningful threshold for defining delayed initiation of adjuvant chemotherapy [23]. For the PDS group, progression-free survival (PFS) was defined as the time from the date of surgery to the first clinical and/or radiological recurrence or progression of the disease. Overall survival (OS) was defined as the time from surgery to death from any cause, or the last follow-up time in patients who were still alive. In the IDS group, the date of diagnosis was used as the starting point for survival analyses, as treatment in this group begins with neoadjuvant chemotherapy, a widely used approach in the literature.

Statistical analysis

All statistical analyses were performed using R software, Version 4.4.2; R Foundation for Statistical Computing, Vienna, Austria. The normality of continuous variables was assessed using visual methods (histograms and Q–Q plots) and the Shapiro–Wilk test.

Continuous variables were expressed as median and interquartile range (IQR), as most variables did not demonstrate normal distribution. Comparisons between two independent groups were performed using the Mann–Whitney U test. Categorical variables were summarized as frequencies and percentages and were compared using Pearson’s chi-square test or Fisher’s exact test when expected cell counts were fewer than five.

Univariable logistic regression analyses were conducted to identify factors associated with postoperative complications. Variables with clinical relevance or statistical significance in univariable analysis were included in the multivariable logistic regression model. Results were reported as odds ratios (ORs) with 95% confidence intervals (CIs). Model fit was assessed using appropriate goodness-of-fit statistics (e.g., Hosmer–Lemeshow test). A two-sided p-value < 0.05 was considered statistically significant.

Survival outcomes, including progression-free survival and overall survival, were analyzed using the Kaplan–Meier method. Patients without an event were censored at the date of last follow-up. Survival curves were compared using the log-rank test, and median survival times with corresponding 95% confidence intervals were reported.

Results

A total of 180 patients were included in the analysis, of whom 20 (11.1%) developed postoperative complications (Table 1). Baseline demographic, clinical, and laboratory characteristics were comparable between patients with and without complications.In contrast, bowel surgery was significantly more frequent among patients who developed complications compared with those who did not (45.0% vs. 18.8%, p = .017). Perioperative transfusion (80.0% vs. 43.1%, p = .002) and postoperative transfusion (70.0% vs. 42.4%, p = .030) were also significantly associated with complications. Anterior peritonectomy was more common in the complication group (35.0% vs. 14.5%), reaching borderline statistical significance (p= .050). Furthermore, higher surgical complexity, as reflected by the Aletti score, was significantly associated with postoperative complications (p= .014), with patients in the highest category (score 3) demonstrating a markedly greater proportion of complications.

Table 1.

Baseline and perioperative characteristics

Variable	No complication (n = 160)	With complication (n = 20)	p-value
Age (years)	59.00 [52.00, 66.25]	64.50 [55.50, 70.25]	0.135
BMI (kg/m²)	27.50 [24.50, 31.60]	25.70 [23.40, 29.60]	0.219
Operation time (min)	240.00 [190.00, 321.25]	275.00 [210.00, 351.25]	0.212
Hemoglobin (g/dL)	11.30 [10.30, 12.25]	11.40 [10.25, 12.03]	0.694
White blood cell count (×10⁹/L)	6.28 [4.86, 8.12]	6.59 [5.52, 7.67]	0.464
Platelet count (×10⁹/L)	268.00 [209.75, 332.75]	207.00 [180.50, 319.50]	0.140
Lymphocyte count (×10⁹/L)	1.80 [1.33, 2.26]	1.80 [1.52, 2.38]	0.594
Neutrophil count (×10⁹/L)	3.50 [2.62, 5.11]	4.48 [3.34, 5.38]	0.290
Albumin (g/L)	42.00 [39.00, 44.00]	41.00 [36.00, 44.00]	0.121
CA125 (U/ml)	55.80 [21.50, 268.00]	128.00 [35.00, 564.00]	0.205
Tumor size (cm)	5.00 [2.50, 10.62]	4.75 [1.88, 8.00]	0.271
Menopause
no	28 (17.5)	1 (5.0)	0.206
yes	132 (82.5)	19 (95.0)
Parity
no	11 (7.6)	0 (0.0)	0.608
yes	134 (92.4)	17 (100.0)
Diabetes Mellitus
no	117 (74.1)	14 (73.7)	1.000
yes	41 (25.9)	5 (26.3)
Hypertension
no	97 (61.4)	10 (52.6)	0.468
yes	61 (38.6)	9 (47.4)
Previous abdominal surgery
no	98 (72.6)	12 (66.7)	0.820
Pfannenstiel	29 (21.5)	5 (27.8)
Midline laparotomy	8 (5.9)	1 (5.6)
Surgery type
PDS	55 (34.4)	7 (35.0)	1.000
IDS	105 (65.6)	13 (65.0)
ASA
ASA 1–2	84 (54.5)	11 (55.0)	1.000
ASA 3–4	70 (45.5)	9 (45.0)
HIPEC
no	148 (92.5)	18 (90.0)	0.657
yes	12 (7.5)	2 (10.0)
Ascites
no	127 (79.4)	16 (80.0)	1.000
yes	33 (20.6)	4 (20.0)
Bowel surgery
no	130 (81.2)	11 (55.0)	0.017
yes	30 (18.8)	9 (45.0)
Skin closure technique
stapler	91 (59.1)	10 (50.0)	0.672
Polypropylene	39 (25.3)	6 (30.0)
stapler+polypropylene	24 (15.6)	4 (20.0)
Suboptimal cytoreduction
no	150 (93.8)	18 (90.0)	0.626
yes	10 (6.2)	2 (10.0)
Perioperative blood transfusion
no	91 (56.9)	4 (20.0)	0.002
yes	69 (43.1)	16 (80.0)
Postoperative blood transfusion
no	91 (57.6)	6 (30.0)	0.030
yes	67 (42.4)	14 (70.0)
Pleural effusion
no	127 (81.4)	14 (70.0)	0.239
yes	29 (18.6)	6 (30.0)
Omental cake
no	61 (39.1)	5 (25.0)	0.327
yes	95 (60.9)	15 (75.0)
Peritoneal carcinomatosis
no	55 (35.3)	6 (30.0)	0.804
yes	101 (64.7)	14 (70.0)
Anterior peritonectomy
no	136 (85.5)	13 (65.0)	0.050
yes	23 (14.5)	7 (35.0)
Pelvic peritonectomy
no	127 (79.4)	12 (60.0)	0.085
yes	33 (20.6)	8 (40.0)
Diaphragmatic peritonectomy
no	128 (80.0)	14 (70.0)	0.381
yes	32 (20.0)	6 (30.0)
Aletti score
1	22 (14.1)	2 (10.0)	0.014
2	118 (75.6)	11 (55.0)
3	16 (10.3)	7 (35.0)
FIGO stage
3	103 (64.4)	11 (55.0)	0.464
4	57 (35.6)	9 (45.0)

Continuous variables are presented as median [interquartile range] and were compared using the Mann–Whitney U test. Categorical variables are presented as n (%) and were compared using the chi-square test or Fisher’s exact test, as appropriate. p < .05 was considered statistically significant

Abbreviations: ASA American Society of Anesthesiologists, BMI Body mass index, FIGO International Federation of Gynecology and Obstetrics, HIPEC Hyperthermic intraperitoneal chemotherapy

In univariable logistic regression analysis, bowel surgery was significantly associated with an increased risk of postoperative complications (OR = 3.55, 95% CI [1.32, 9.34], p = .010) (Table 2). Similarly, perioperative transfusion was strongly associated with complications (OR = 5.28, 95% CI [1.84, 19.05], p = .004), as was postoperative transfusion (OR = 3.17, 95% CI [1.20, 9.34], p = .025). Anterior peritonectomy was also identified as a significant predictor (OR = 3.18, 95% CI [1.10, 8.67], p = .026).

Table 2.

Univariable logistic regression for complications

Variable	OR (95% CI)	p-value
Age (years)	1.03 (0.99–1.08)	0.193
BMI (kg/m²)	0.94 (0.85–1.02)	0.171
Operation time (min)	1 (1–1.01)	0.333
Hemoglobin (g/dL)	0.92 (0.69–1.27)	0.603
White blood cell count (×10⁹/L)	1.01 (0.83–1.18)	0.933
Platelet count (×10⁹/L)	1 (0.99–1)	0.363
Lymphocyte count (×10⁹/L)	1.22 (0.62–2.34)	0.549
Neutrophil count (×10⁹/L)	1.02 (0.83–1.21)	0.803
Albumin (g/L)	0.92 (0.85–1.01)	0.077
CA125 (U/ml)	1 (1–1)	0.639
Tumor size (cm)	0.94 (0.85–1.02)	0.179
Menopause	4.03 (0.78–73.89)	0.183
Diabetes Mellitus	1.02 (0.31–2.85)	0.973
Hypertension	1.43 (0.54–3.75)	0.462
Previous surgery	1.17 (0.48–2.48)	0.705
Surgery type	0.97 (0.38–2.72)	0.956
ASA (ASA 3–4)	0.98 (0.38–2.51)	0.969
HIPEC	1.37 (0.2–5.57)	0.695
Ascites	0.96 (0.26–2.83)	0.948
Bowel surgery	3.55 (1.32–9.34)	0.010
Skin closure	1.26 (0.68–2.24)	0.451
Suboptimal cytoreduction	1.67 (0.24–6.98)	0.530
Perioperative blood transfusion	5.28 (1.84–19.05)	0.004
Postoperative blood transfusion	3.17 (1.2–9.34)	0.025
Pleural effusion	1.88 (0.62–5.12)	0.234
Omental cake	1.93 (0.71–6.17)	0.226
Peritoneal carcinomatosis	1.27 (0.48–3.76)	0.642
Anterior peritonectomy	3.18 (1.1–8.67)	0.026
Pelvic peritonectomy	2.57 (0.94–6.73)	0.058
Diaphragmatic peritonectomy	1.71 (0.57–4.64)	0.306
Aletti score 2	1.03 (0.25–6.92)	0.975
Aletti score 3	4.81 (1.01–35.29)	0.070
FIGO stage 4	1.48 (0.57–3.78)	0.414

Abbreviations: ASA American Society of Anesthesiologists, BMI Body mass index, FIGO International Federation of Gynecology and Obstetrics, HIPEC Hyperthermic intraperitoneal chemotherapy

Pelvic peritonectomy demonstrated borderline statistical significance (OR = 2.57, 95% CI [0.94, 6.73], p = .058). Among surgical complexity categories, patients with Aletti score 3 had an elevated risk of complications compared to the reference category (OR = 4.81, 95% CI [1.01, 35.29]), although this did not reach conventional statistical significance (p = .070). Albumin level showed a borderline protective association (OR = 0.92, 95% CI [0.85, 1.01], p = .077). No other variables were significantly associated with postoperative complications (Table 2).

In the multivariable logistic regression model including bowel surgery, perioperative transfusion, and anterior peritonectomy, only perioperative transfusion remained independently associated with postoperative complications (Table 3). Patients who received perioperative transfusion had a significantly increased risk of complications (OR = 3.72, 95% CI [1.17, 14.25], p = .035). Bowel surgery was no longer significantly associated with complications after adjustment (OR = 1.31, 95% CI [0.81, 2.07], p = .257). Similarly, anterior peritonectomy did not retain statistical significance in the adjusted model (OR = 2.02, 95% CI [0.66, 5.82], p = .202). Overall, after controlling for surgical factors, perioperative transfusion emerged as the only independent predictor of postoperative complications.

Table 3.

Multivariable logistic regression

Variable	OR (95% CI)	p-value
Bowel surgery	1.31 (0.81–2.07)	0.257
Perioperative blood transfusion	3.72 (1.17–14.25)	0.035
Anterior peritonectomy	2.02 (0.66–5.82)	0.202

Major postoperative complications (Clavien–Dindo grade ≥ III) are summarized in Table 4. Intraabdominal abscess was the most common complication. Other complications included intraabdominal bleeding, intestinal perforation, anastomotic leakage, relaparotomy for fascial dehiscence, pulmonary embolism requiring ICU admission, and acute kidney injury. Two patients died within 30 days after surgery.

Table 4.

Major postoperative complications (Clavien–Dindo ≥ III)

Complication Type	Clavien–Dindo Grade	n (%)
Intraabdominal bleeding	IIIb	1 (0.6)
Anastomotic leak	IIIb	2 (1.1)
Intraabdominal abscess	IIIa IIIb	4 (2.2) 4 (2.2)
Intestinal perforation	IIIb	1 (0.6)
Relaparotomy for fascial dehiscence	IIIb	3 (1.7)
Pulmonary embolism requiring ICU admission	IVa	2 (1.1)
Acute kidney injury	IVa	1 (0.6)
30-day mortality	V	2 (1.1)

Abbreviations: ICU Intensive care unit

Patients were categorized into early (≤ 6 weeks) and delayed (> 6 weeks) chemotherapy groups (Table 5). Overall, 70 patients (48.6%) initiated adjuvant chemotherapy in the early group, whereas 74 patients (51.4%) experienced delayed initiation. The median time to chemotherapy initiation was 34.5 days (IQR 30.2–39.0) in the early (≤6 weeks) group and 57.0 days (IQR 51.0–65.8) in the delayed (>6 weeks) group. Baseline demographic, operative, and laboratory characteristics were comparable between the early and delayed chemotherapy groups, except for platelet count. Platelet levels were significantly higher in the delayed group compared with the early group (median 282 [IQR 211–354] vs. 244 [IQR 182–318], p = 0.022).

Table 5.

Comparison of clinical and laboratory characteristics between early (≤ 6 weeks) and delayed (> 6 weeks) chemotherapy groups

Variable	Early (≤ 6 weeks) (n = 70)	Delayed (> 6 weeks) (n = 74)	p-value
Age (years)	60 [53–67]	60 [51–66.8]	0.596
BMI (kg/m²)	27.0 [24.5–30.4]	27.3 [24.3–31.5]	0.651
Operation time (min)	250 [193.8–320]	250 [198.8–355]	0.555
Hospital stay (days)	8 [7–11]	8 [7–10]	0.925
Hemoglobin (g/dL)	11.2 [10.4–11.8]	11.4 [10.2–12.4]	0.277
White blood cell count (×10⁹/L)	6.35 [4.61–7.68]	6.24 [5.07–7.52]	0.570
Platelet count (×10⁹/L)	244 [182–318]	282 [211–354]	0.022
Lymphocyte count (×10⁹/L)	1.72 [1.22–2.34]	1.79 [1.33–2.13]	0.842
Neutrophil count (×10⁹/L)	3.36 [2.29–4.92]	3.84 [2.90–4.98]	0.121
Albumin (g/L)	42 [40–45]	42 [36.8–44]	0.103
CA125 (U/ml)	68.8 [17.3-252.5]	44.9 [24.8–391]	0.627
Tumor size (cm)	5 [3–9]	5.5 [3–10.4]	0.420

Categorical clinical and surgical characteristics were compared between early (≤ 6 weeks) and delayed (> 6 weeks) chemotherapy groups (Table 6). Overall, baseline clinical and operative characteristics were comparable between groups. Surgery type (IDS) was significantly more frequent in the early chemotherapy group compared with the delayed group (80.0% vs. 55.4%,p= .003). Omental cake (72.1% vs. 55.4%, p = .060) and peritoneal carcinomatosis (75.0% vs. 59.5%, p = .074) showed borderline statistical significance, with higher proportions observed in the early chemotherapy group. Overall, aside from surgery type, baseline clinical and operative characteristics were comparable between early and delayed chemotherapy groups.

Table 6.

Comparison of categorical clinical and surgical characteristics between early (≤ 6 weeks) and delayed (> 6 weeks) chemotherapy groups

Variable	Early (≤ 6 weeks) (n = 70)	Delayed (> 6 weeks) (n = 74)	p-value
Complication (Yes)	5 (7.2)	12 (16.4)	0.153
Menopause (Yes)	62 (88.6)	60 (81.1)	0.309
Parity			0.088
no	2 (3.3)	9 (13.2)
yes	59 (96.7)	59 (86.8)
Diabetes mellitus	17 (24.6)	20 (27.4)	0.855
Hypertension	23 (33.3)	28 (38.4)	0.654
Previous surgery			0.585
None	41 (70.7)	43 (67.2)
Pfannenstiel	15 (25.9)	16 (25.0)
Midline laparotomy	2 (3.4)	5 (7.8)
Surgery type			0.003
IDS	56 (80.0)	41 (55.4)
PDS	14 (20.0)	33 (44.6)
ASA score			0.740
1	1 (1.5)	2 (2.7)
2	36 (54.5)	37 (50.7)
3	29 (43.9)	33 (45.2)
4	0 (0.0)	1 (1.4)
Aletti score			0.150
low	14 (20.6)	7 (9.5)
intermediate	43 (63.2)	56 (75.7)
high	11 (16.2)	11 (14.9)
FIGO stage			0.799
3	43 (61.4)	48 (64.9)
4	27 (38.6)	26 (35.1)
HIPEC	5 (7.1)	7 (9.5)	0.841
Ascites	16 (22.9)	12 (16.2)	0.426
Bowel surgery			0.838
None	54 (77.1)	57 (77.0)
anastomosis	8 (11.4)	6 (8.1)
ostomy	5 (7.1)	6 (8.1)
anastomosis+ ostomy	3 (4.3)	5 (6.8)
Perioperative blood transfusion	33 (47.1)	38 (51.4)	0.735
Postoperative blood transfusion	32 (47.1)	34 (45.9)	1.000
ICU admission	48 (68.6)	51 (68.9)	1.000
Suboptimal cytoreduction	4 (5.7)	3 (4.1)	0.940
Pleural effusion	13 (19.4)	13 (17.8)	0.980
Omental cake	49 (72.1)	41 (55.4)	0.060
Peritoneal carcinomatosis	51 (75.0)	44 (59.5)	0.074
Anterior peritonectomy	12 (17.1)	13 (17.6)	1.000
Pelvic peritonectomy	18 (25.7)	14 (18.9)	0.436
Diaphragmatic peritonectomy	17 (24.3)	18 (24.3)	1.000

Continuous variables are presented as median [interquartile range] and were compared using the Mann–Whitney U test

Categorical variables are presented as n (%) and were compared using the Chi-square or Fisher’s exact test as appropriate

Bold values indicate statistical significance (p < .05)

Abbreviations: ASA American Society of Anesthesiologists, BMI Body mass index, ICU Intensive care unit, FIGO International Federation of Gynecology and Obstetrics, HIPEC Hyperthermic intraperitoneal chemotherapy

Forest plot of the multivariable logistic regression analysis identifying independent predictors of postoperative complications (Fig. 1). Odds ratios (ORs) with 95% confidence intervals (CIs) are presented on a logarithmic scale. The dashed vertical line indicates the null value (OR = 1). Perioperative transfusion was independently associated with increased odds of complications, whereas bowel surgery category and anterior peritonectomy were not statistically significant in the adjusted model.

Fig. 1.

Forest plot of multivariable logistic regression analysis for postoperative complications

Survival analyses were performed in patients with available survival data. Patients whose survival data could not be obtained because follow-up continued at other institutions were excluded.

Median follow-up time was 32 months for PDS patients and 28 months for IDS patients, estimated using the reverse Kaplan–Meier method (Fig. 2). Kaplan–Meier survival analyses were conducted to evaluate the effect of treatment initiation timing (early, ≤6 weeks vs. delayed >6 weeks) on PFS and OS.

Fig. 2.

Kaplan–Meier curves showing progression-free survival and overall survival according to timing of chemotherapy initiation in PDS and IDS groups (A) PFS in the PDS patients; B PFS in the IDS patients; C OS in the PDS patients; and D OS in the IDS patients

In PDS patients, no statistically significant difference was observed between the early and delayed treatment groups in PFS (p = .11; Fig. 2A). Similarly, in IDS patients, the PFS comparison did not reach statistical significance, although a trend toward significance was observed (p = .072; Fig. 2B). For OS, no significant difference was observed between early and delayed treatment in PDS patients (p = .55; Fig. 2C). In contrast, among IDS patients, the early treatment group demonstrated significantly improved overall survival compared with the delayed group (p = .0013; Fig. 2D). These findings suggest that early treatment initiation may confer a significant survival advantage in IDS patients. Although statistical significance was not achieved for PFS outcomes, a trend favoring early treatment was observed in the IDS group.

Kaplan–Meier analysis demonstrated a statistically significant difference in OS between patients undergoing primary debulking surgery and interval debulking surgery (log-rank p = 0.021), with improved survival observed in the PDS group. In contrast, no statistically significant difference was found in PFS between the two groups (log-rank p = 0.17), although a trend toward improved PFS was noted in the PDS group (Fig. 3).

Fig. 3.

Kaplan–Meier survival curves comparing patients undergoing primary debulking surgery and interval debulking surgery (A) Progression-free survival; B Overall survival

Discussion

In our study, 65.6% of patients underwent IDS, whereas 34.4% underwent PDS. The relatively high proportion of patients undergoing IDS may reflect the advanced disease burden and the patient population characteristic of a tertiary referral center. The rate of postoperative complications after cytoreductive surgery was found to be 11.1% in patients with advanced-stage epithelial ovarian cancer. In our comparative analyses, bowel surgery during the operation and perioperative and postoperative blood transfusions were found to be significant in the group that developed complications. Multivariable analyses showed that perioperative transfusion was an independent risk factor for postoperative complications. When the timing of adjuvant chemotherapy initiation was examined, patients who underwent IDS were found to be able to start chemotherapy earlier than those in the PDS group. In addition, when the timing of adjuvant chemotherapy initiation was evaluated, early treatment initiation was significantly associated with overall survival, especially in patients who underwent IDS.

Because complete cytoreduction is targeted in advanced-stage ovarian cancer surgery, complex surgical procedures are performed [7]. These surgical procedures increase the risk of complications for patients. In our study, the CDC classification was used for objective evaluation. As the CDC grade increases, hospital stay becomes longer and may result in outcomes such as extended intensive care unit stay and death [24]. In the literature, the rate of CDC grade IIIa or higher complications ranges from 10% to 25% [25–28]. Similarly, in our study, the complication rate was 11.1%. This result suggests that the surgical approach at our center is within acceptable morbidity limits and that our results are consistent with the literature.

When factors associated with postoperative complications were evaluated in our study, surgery-related variables appeared to be more prominent and determinative in their development. Studies show that increased surgical complexity is associated with a higher risk of postoperative complications [28, 29]. A 2022 meta-analysis also showed that surgical complexity was associated with an increased risk of complications [30]. In our study, the Aletti scoring system was used to evaluate the degree of surgical complexity. Although higher surgical complexity was observed in the group that developed complications in comparative analyses, this association did not remain independent in multivariable analysis. This result suggests that the effect of surgical complexity on the development of complications may be indirect. In the study by Castro et al., no significant difference in complication rates was found between the PDS and IDS groups, and they reported that the most important factors in the development of complications were comorbidities and intraoperative bleeding [23]. Bianchi et al. showed that multiple bowel resections and intraoperative blood loss were independent risk factors for complications after cytoreductive surgery in patients with advanced-stage ovarian cancer [31]. In our study as well, bowel resection during surgery and perioperative transfusion were found to be associated with complications. However, in multivariable analysis, only perioperative transfusion remained an independent predictor. This result suggests that increased blood loss during surgery and the resulting need for transfusion may play an important role in the development of postoperative complications.

There are studies showing that delayed initiation of adjuvant chemotherapy after surgery in ovarian cancer has negative effects on overall survival [32, 33]. However, it has also been reported that starting chemotherapy very early does not provide an additional survival benefit [34]. Although initiation of adjuvant chemotherapy within 3–4 weeks is considered ideal, a 6-week threshold has been used in the literature as a clinically meaningful upper limit [18, 23, 35]. This timeframe reflects a balance between adequate postoperative recovery and avoidance of clinically significant treatment delay. In addition, the FRANCOGYN study, which evaluated patients undergoing primary complete macroscopic cytoreduction, reported a median time to chemotherapy of 43 days. These findings support the use of a 6-week threshold as a clinically meaningful and pragmatic definition of delayed treatment [36]. In our study, patient- and surgery-related factors may affect the time to initiation of adjuvant chemotherapy were evaluated, but no significant associations were found for most variables. In contrast, when the surgery type was examined, patients who underwent interval debulking surgery were found to be able to start chemotherapy earlier. Earlier initiation of chemotherapy in patients who underwent IDS may be related to lower surgical complexity, reduced intraoperative blood loss, and a faster postoperative recovery process due to lessened tumor burden after neoadjuvant treatment. In addition, organizational factors, such as the timing of histopathological evaluation and the scheduling of oncologic treatments, may also have contributed to the earlier initiation of chemotherapy in this group.

Although several studies in the literature have shown that postoperative complications delay the initiation of adjuvant chemotherapy [23, 27], this association was not significant in our study. This may be related to effective management of complications and differences in patient selection. Indeed, some studies have reported that when complications are managed appropriately, the time to initiation of adjuvant chemotherapy may not be affected [28]. This finding suggests that delayed initiation of chemotherapy in our study was not directly attributable to postoperative complications and that other factors may also have played a role, while effective management of complications may help prevent delays in chemotherapy.

In survival analyses, the effect of adjuvant chemotherapy timing differed by surgical group. While no significant difference was found between early and delayed chemotherapy initiation in terms of progression-free survival and overall survival in the PDS group, overall survival was significantly better in IDS patients who started treatment earlier. Consistent with our findings, Steventon et al. found that chemotherapy initiated more than 6 weeks after diagnosis was associated with worse overall survival, particularly in the IDS group, whereas no significant association was observed in the PDS group [35]. A multicenter study published in 2023 also showed that delayed chemotherapy initiation negatively affected OS but had no significant effect on PFS in patients who underwent IDS after NACT [37]. These results indicate that maintaining treatment continuity may be more critical from a prognostic perspective in patients operated on after neoadjuvant treatment. Although statistical significance was not reached for progression-free survival in the IDS group, a trend in favor of early treatment supports this relationship. When survival outcomes between PDS and IDS are compared, the literature shows heterogeneous findings; while some studies report comparable survival between the two approaches, others have demonstrated a survival advantage for PDS, particularly in patients in whom complete cytoreduction is achieved [38–40]. In our study, the finding of improved overall survival in the PDS group is consistent with studies suggesting that selected patients undergoing successful PDS may derive a survival benefit.

Conclusion

In conclusion, our study showed that morbidity developing after cytoreductive surgery in patients with advanced-stage epithelial ovarian cancer is particularly associated with surgery-related factors, and that the effect of the timing of adjuvant chemotherapy initiation on survival may differ according to surgical subgroups. Timely initiation of chemotherapy appears to be more critical for overall survival, especially in patients who underwent IDS. Therefore, reducing perioperative morbidity and maintaining treatment continuity should be considered important goals in patient management.

Strengths and limitations

One of the strengths of this study is that it was conducted in a high-volume center accredited by ESGO, which is important for showing a relatively standardized management process across surgical approach, perioperative care, and adjuvant treatment planning. The fact that the study evaluated not only postoperative complications but also the time to initiation of adjuvant chemotherapy and its relationship with survival outcomes, provides a clinically comprehensive assessment. Another strength is that the PDS and IDS groups were evaluated separately; this allowed demonstration that the prognostic effect of chemotherapy timing may differ across surgical subgroups.

The main limitations of the study are its retrospective design, its single-center setting, and the limited number of patients, particularly in subgroup analyses. Showing a single-center experience may limit the generalizability of the findings to different centers and patient populations. Important prognostic factors such as chemotherapy regimen, BRCA and HRD status, and the use of maintenance therapy were not available in our dataset and should be considered when interpreting our results. The number of patients undergoing HIPEC in our cohort was limited, and therefore no subgroup analysis was performed regarding its impact on outcomes.

Abbreviation

ASA

American Society of Anesthesiologists classification

CDC

Clavien–Dindo classification

CIs

Confidence intervals

ICU

Intensive Care Unit

IDS

Interval debulking surgery

IQR

Interquartile range

ECOG

Eastern Cooperative Oncology Group

ESGO

European Society of Gynaecological Oncology

FIGO

International Federation of Obstetrics and Gynaecology

HIPEC

Hyperthermic intraperitoneal chemotherapy

NACT

Neoadjuvant chemotherapy

NCCN

National Comprehensive Cancer Network

PDS

Primary debulking surgery

PFS

Progression-free survival

STROBE

Strengthening the Reporting of Observational Studies in Epidemiology

ORs

Odds ratios

OS

Overall survival

Authors’ contributions

Conceptualization: NÇÇ , GŞ, İTYData curation: NÇÇ, MY, YÖÖFormal analysis: NÇÇ , MY , YÖÖInvestigation: NÇÇ, GŞ, İTYMethodology: NÇÇ, İTYSoftware: NÇÇ, MY, YÖÖSupervision: NÇÇ , GŞ, İTYValidation: NÇÇ , MYVisualization: NÇÇ, YÖÖWriting - original draft: NÇÇ, MY, YÖÖWriting - review & editing: NÇÇ, GŞ, İTY.

Funding

The authors received no specific funding for this work.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Başakşehir Çam and Sakura City Hospital (Approval number: 96317027-514.10-287954128). According to Turkish legislation and the institutional ethics committee policies of our hospital, informed consent is not required for retrospective studies.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.