- Email: [email protected]
Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interval debulking surgery
YGYNO-976897; No. of pages: 5; 4C: Gynecologic Oncology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interval debulking surgery Kristen Pepin a,b,⁎, Amy Bregar a, Michelle Davis b, Alexander Melamed a, Emily Hinchcliff c, Allison Gockley b, Neil Horowitz b, Marcela G. del Carmen a a b c
Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114, United States Brigham and Women's Hospital, 75 Francis St, Boston, MA 20115, United States MD Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, United States
H I G H L I G H T S • • • •
Intensive care unit (ICU) admissions are more common in ovarian cancer patients treated with primary debulking (PDS). This was true though patients treated with neoadjuvant chemotherapy (NACT) were older and had more advanced disease. NACT patients admitted to the ICU had comparable surgical complexity at time of cytoreduction as PDS ICU admitted patients. Indications of ICU admission were not different between patients undergoing PDS vs NACT.
a r t i c l e
i n f o
Article history: Received 13 July 2017 Received in revised form 22 September 2017 Accepted 24 September 2017 Available online xxxx Keywords: Intensive care Cytoreduction Neoadjuvant Surgical complexity
a b s t r a c t Objective. Admissions to intensive care units (ICU) are costly, but are necessary for some patients undergoing radical cancer surgery. When compared to primary debulking surgery (PDS), neoadjuvant chemotherapy (NACT) with interval debulking surgery, is associated with less peri-operative morbidity. In this study, we compare rates, indications and lengths of ICU stays among ovarian cancer patients admitted to the ICU within 30 days of cytoreduction, either primary or interval. Methods. A retrospective chart review was performed of patients with stage III-IV ovarian cancer who underwent surgical cytoreduction at two large academic medical centers between 2010 and 2014. Chi square tests, Student t-tests, and Mann-U Whitney tests were used. Results. A total of 635 patients were included in the study. There were 43 ICU admissions, 7% of patients. Compared to NACT, a higher percentage of PDS patients required ICU admission, 9.4% vs 3.9% of patients (P = 0.004). ICU admission indications did not vary between PDS and NACT patients. NACT patients admitted to the ICU had comparable mean surgical complexity scores to those PDS patients admitted to the ICU, 6.2 (95%CI 5.3–7.1) vs 4.5 (95%CI 3.1–6.0) (P = 0.006). Length of ICU admission did not vary between groups, PDS 2.7 days (95%CI 2.3–3.2) vs 3.5 days (95%CI 1.5–5.6) for NACT (P = 0.936). Conclusions. The rate of ICU admissions among patients undergoing PDS is higher than for NACT. Among patients admitted to the ICU, indications for admission, length of stay and surgical complexity were similar between patients treated with NACT and PDS. © 2017 Elsevier Inc. All rights reserved.
1. Introduction Admission to an intensive care unit (ICU) represents a significant cost to the health care system [1]. It has been estimated to account for over half a percent of the United States Gross Domestic Product (GDP) ⁎ Corresponding author at: 89 East Brookline St, Apt 4, Boston, MA 02115, United States. E-mail address: [email protected] (K. Pepin).
[1]. Given the lack of screening for detection of early-stage ovarian cancer, an estimated 85% of patients with epithelial ovarian cancer present with advanced-stage (III/IV) disease and require radical debulking surgery, either upfront or as an interval strategy, which often necessitates ICU-level of care. Post-operative ICU admission rates after ovarian cancer debulking surgery have been reported to be as high as 30% [2]. Predictors of extended ICU stay in these patients include older age, more medical comorbidities and more extensive surgical debulking [3].
https://doi.org/10.1016/j.ygyno.2017.09.028 0090-8258/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028
2
K. Pepin et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Since the publication of two phase III randomized clinical trials, Kehoe et al. and Vergote et al., showing equivalent survival between ovarian cancer patients treated with primary debulking surgery (PDS) and neoadjuvant chemotherapy with interval debulking surgery (NACT), the use of NACT has increased [4–6]. NACT has become a preferred paradigm for the management of patients with unresectable disease upfront or those too medically infirmed to undergo PDS. This is despite the relatively low rates of resection to no residual disease (R0) and lower than previously reported survival, in these trials [5–6]. Though poor candidates for upfront surgery, were historically offered NACT, the before mentioned non-inferiority trials, have likely resulted in a lower the threshold to offer patients NACT. Several studies have reported fewer post-operative complications and lower 30-day mortality among patients treated with NACT vs PDS [5–6]. With the recent shift in treatment of older and more medically complex women with NACT, there have been no accompanying data on how ICU utilization has been affected. In this study, we compare rates, indications and lengths of ICU stays among ovarian cancer patients admitted to the ICU within 30 days of cytoreduction, either primary or interval. 2. Methods We identified all newly diagnosed stage III-IV epithelial ovarian, fallopian tube and primary peritoneal cancer patients who underwent surgical cytoreduction, at two large academic medical centers between 2010 and 2014. A joint institution Institutional Review Board (IRB) approved the retrospective chart review and analysis. Patients were identified using tumor registries and tumor board records. Patients with early stage tumors, germ cell tumors and borderline tumors were not included in the analysis. Patients who were treated exclusively with chemotherapy were excluded from the study, as were patients who underwent debulking surgery at an outside facility. Patients were defined as having had primary debulking surgery if surgery was the first treatment modality the patient underwent, and an attempt at debulking was made at the time of surgery. Patients who received only surgery, and no adjuvant chemotherapy were considered part of the PDS group. The NACT group was defined by any chemotherapy prior to attempt at surgical debulking. In most cases, the planned course was for a debulking procedure following 3 cycles of NACT. The Charlson Comorbidity Index was used to evaluate preoperative comorbid conditions at the time of diagnosis. The index includes predicts 10-year mortality, based on history of myocardial infarction, congestive heart failure, renal disease, diabetes, chronic obstructive pulmonary disease, liver disease, dementia, peripheral vascular disease, peptic ulcer disease, hemiplegia and malignancy. All patients in the study, by definition, had metastatic cancer, earning a Charlson Comorbidity score of 6. For this analysis, their scores were calculated excluding their cancer diagnosis, such that the lowest possible score was zero [7]. Additionally, no condition included in the Comorbidity index should have been significantly improved by use of chemotherapy, thus scores were not recalculated after completion of NACT. Surgical morbidity was characterized by the surgical procedures performed at the time of cytoreduction, including bowel resection, ostomy creation, splenectomy, liver resection and procedures on the diaphragm. All procedures performed were incorporated into a surgical complexity score, to compare the extent of surgical debulking [8]. One point was assigned for hysterectomy/bilateral salpingo-oophorectomy, omentectomy, pelvic lymph node dissection, paraaortic lymph node dissection and small bowel resection. Two points were assigned for large bowel resection, diaphragm stripping/resection, splenectomy and liver resection. Three points were assigned for a rectosigmoid resection with anastomosis. Post-operative morbidity was defined by estimated blood loss (EBL) N 2 l, intensive care unit (ICU) admission, need for re-operation, hospital length of stay (LOS), and readmission within 30-days.
Descriptive statistics were defined with means and medians. Chi square tests, Student t-tests, and Mann-U Whitney tests were used to evaluate categorical and continuous variables. All analysis was performed in SPSS. 3. Results A total of 635 patients met inclusion criteria for the study. There were 43 ICU admissions, just 7% of patients who underwent surgical debulking during this time period. The median age was 64 (range 27–93) and median BMI 25.5 (range 16.1–72.1). Of the included patients, 459 (72%) had stage III disease and 176 (28%) had stage IV disease. Age, race, BMI, diagnosis CA 125 (including CA 125 prior to initiation of chemotherapy for NACT patients), comorbidity index and stage did not vary significantly between patients who were and were not admitted to an ICU post-operatively (Table 1). Those undergoing NACT were older, 64.1 years (range 34–89), than those undergoing PDS, 62 (range 27–93) (P = 0.008) and were more likely to have stage IV disease; 42% of NACT patients vs 14% PDS patients (P = 0.000). Comorbidity index did not vary significantly between the two surgical groups. Rates of NACT rose overtime during the study period (P = 0.001) (Fig. 1). Among the NACT patients, b1% received b 3 cycles of preoperative chemotherapy, 55% received three cycles of preoperative chemotherapy, 28% received four cycles of preoperative chemotherapy and the remainder received N4 cycles of preoperative chemotherapy. Three patients treated with neoadjuvant chemotherapy underwent a laparoscopic debulking procedure, with the rest done via laparotomy. The most common indications for ICU admission were hemodynamic monitoring/pharmacologic blood pressure support (84.1%), respiratory failure/inability to extubate (22.7%), bacteremia/sepsis (18.2%), anastomotic leak (9.1%), cardiac indication (4.6%), neurologic indication (4.5%) and open abdomen (4.5%). Fourteen patients had more than one indication for ICU admission. Patients requiring ICU admission had higher estimated blood loss; 1630 ml (95%CI 1181–2147) vs 503 ml (95%CI 453–540), (P b 0.001). ICU admission indications did not vary between PDS and NACT patients (Fig. 2). Patients admitted to a ICU had significantly higher pre-operative CA − 125 levels; 4125 U/ml (95%CI 702–11,135) vs 545 U/ml (95%CI 435–678), (P b 0.001). Most patients were admitted to the ICU during the initial hospitalization,
Table 1 Patient characteristics. ICU admitted
Age BMI Dx CA 125
Non-ICU admitted
P value
Mean
Range
Mean
Range
64 26.8 6080
39–79 19.0–45.9 4–37,600
63 26.5 1445
26–93 16.1–72.1 22–103,525
Race White Black Hispanic Asian Other/Pt declined Stage 3 4 Histology Serous Other Comorbidity index 0 1 2 3 4 or greater Neoadjuvant chemotherapy
Number
%
Number
%
26 2 0 3 8
61% 5% 0% 7% 27%
470 14 8 18 56
80% 2% 1% 3% 14%
430 162
73% 27%
29 14
67% 33%
387 205
65% 35%
27 16
63% 26%
62 128 166 135 101 12
10% 22% 28% 23% 17% 28%
2 9 15 7 10 294
5% 21% 35% 16% 23% 50%
0.547 0.565 0.267 P value 0.232
0.283
0.952
0.204
0.000
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028
K. Pepin et al. / Gynecologic Oncology xxx (2017) xxx–xxx
3
Table 2 Surgical morbidity. Surgical morbidity
Splenectomy Bowel surgery Liver resection Diaphragm surgery EBL N 2 l Surgical complexity score
Fig. 1. Use of NACT by study year.
but 7% of the ICU admissions occurred during a readmission within 30 days. Compared to NACT, a higher percentage of primary debulking patients required ICU admission, 9.4% vs 3.9% of neoadjuvant chemotherapy patients (P = 0.004). Length of ICU admission did not vary between groups, PDS 2.7 days (95%CI 2.3–3.2) vs 3.5 days (95%CI 1.5–5.6) for NACT (P = 0.936). Total hospital length of stay was longer among PDS vs NACT patients requiring ICU admission, PDS 14.6 days (95%CI 11.3–17.6) vs NACT 9.8 days (95%CI 6.2–13.5), (P = 0.040). Among all patients, median time to first post-operative chemotherapy, though significant varied by just 4 days (PDS = 29 days, 95%CI 28–30 vs NACT = 25 days, 95%CI 24–27, P b 0.001). This was not significantly different for patients who were admitted to an ICU, with a mean time to first postoperative chemotherapy of 27 days (95%CI 21–33). Surgical complexity varied significantly between those admitted and not admitted to the ICU, with splenectomy (35% vs 9%), bowel resection (58%vs 24%), and diaphragmatic surgery (23% vs 10%) being more common in ICU admitted patients (Table 2). Though both groups, ICU admitted and non-ICU admitted had surgical complexity scores ranging 1–11,
ICU admitted
Non-ICU admitted
(n)
% or range
(n)
% or range
15 25 18 10 17 5.7
35% 58% 3% 23% 40% 1–11
52 140 4 59 13 3.4
9% 24% 9% 10% 2% 1–11
P value
b0.001 b0.001 P = 0.054 P = 0.012 b0.001 P = 0.004
those admitted to the ICU had higher mean surgical complexity scores, 5.7 (95%CI 5.0–6.5) vs 3.4 (95%CI 3.2–3.5) (P = 0.006). However, when comparing just those patients admitted to the ICU, rates of splenectomy (39% vs 23%), diaphragmatic surgery (23% vs 23%) and liver resection (10% vs 8%) were similar between those treated with PDS vs NACT (P N 0.05). Patients admitted to the ICU undergoing PDS had higher rates of bowel resection that those undergoing NACT, 71% vs 23% (P = 0.005). Overall surgical complexity scores did not vary significantly between PDS and NACT patients admitted to the ICU, PDS = 6.2 (95%CI 5.3–7.1) vs 4.5 (95%CI 3.1–6.0), (P = 0.600) (Table 3). Patients admitted to the ICU were more likely to have an estimated blood loss N2 l, 40% vs 2% (P b 0.001). Among patients admitted to an ICU, there was not a difference in estimated blood loss between PDS, 1700 cm3 (95%CI 1200–2200) and NACT patients, 1000 cm3 (95%CI 700–2000), (P N 0.50). Patients admitted to an ICU were more likely to be readmitted to the hospital within 30 days (28% vs 13%), undergo reoperation (21% vs b 1%) and to have a postoperative pulmonary embolus (18% vs 3%), as compared with patients not admitted to an ICU (P b 0.01) (Table 4). There were no differences in the rates of these complications between PDS and NACT patients admitted to the ICU; readmission to the hospital within 30 days (29% vs 23%), reoperation (23% vs 15%) and postoperative pulmonary embolus (19% vs 15%), (P N 0.05).
Fig. 2. ICU admission indications by treatment type.
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028
4
K. Pepin et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Table 3 Mean surgical complexity scores. ICU admitted
Non-ICU admitted
PDS
NACT
Mean 95%CI
Mean 95%CI
6.2
5.3–7.1 4.5
P value
PDS
NACT
Mean 95%CI
Mean 95%CI
3.1–6.0 0.600 3.7
3.4–3.9 3.0
P value
2.8–3.2 0.000
4. Conclusions Patients with advanced-stage ovarian cancer often require ICU-level of care. Previously published data have focused on identification of predisposing pre-operative, intraoperative and post-operative factors that increase risk of ICU utilization [9]. However, all literature on ICU admissions among ovarian cancer patients was published during a time when the vast majority of patients were undergoing primary debulking surgery. We analyzed ICU admissions among patients treated between 2010 and 2014, a time during which adoption of NACT increased, following publication of studies showing no difference in overall survival when compared to patient treated by PDS [5–6]. Older age, greater disease burden and more severe medical comorbidities have all been cited as factors predisposing ovarian cancer patients to ICU admission at the time of surgery [3,10]. These may be factors that make patients more likely to be considered for NACT as opposed to PDS. One study showed a 5.9-fold increase in prolonged ICU stay in ovarian cancer patients undergoing debulking surgery over the age of 63 [2]. Patients in our study who received NACT were older and were more likely to have Stage IV disease, though comorbidity index did not vary significantly between groups. The comorbidity index used in this study, while predictive of 10-year mortality, does not capture many conditions that complicate surgical management, including pulmonary embolism, deep venous thrombosis, pleural effusion, morbid obesity, smoking status, immobility, infectious diseases, cardiomyopathy etc. This likely accounts for there being no difference in comorbidity index between PDS and NACT groups, despite NACT generally being intended for more medically complicated patients. The ICU admission rate was significantly lower among patients treated with NACT vs PDS, though NACT patients were on average older and had more advanced disease. This suggests that treating patients with NACT is a way to minimize ICU admissions among the patients most at risk of ICU admission. In our study, 7% of patients undergoing debulking surgery required ICU admission. This rate is on the lower end of what has been previously reported in other studies, with ICU admissions in this patient population reported to range between 6% and 36% [3,9–11]. The rates of ICU admission for patients treated with NACT (3.9%) in our study were significantly lower than those undergoing PDS (9.4%). This finding, may be explained, at least in part, by the fact that prior to the documentation of non-inferiority of NACT when compared to PDS, upfront surgery was performed more frequently and is associated with higher rates of ICU admission. Additionally, most publications on ICU admissions among gynecologic oncology patients focus on only immediate post-operative ICU admissions. In our series, we documented ICU admissions related to Table 4 Post-operative morbidity.
Conflict of interest statement The authors of this manuscript certify that they have no affiliations with or involvement in any organization or entity with a financial interest or non-financial interest in the subject matter discussed in this manuscript.
Post-operative morbidity
Readmit Re-operation DVT/PE Death Hospital length of stay
surgical morbidity, underlying medical comorbidity, and postoperative complications. Thus, we included ICU admissions that occurred within 30 days of surgery. The most common indications for ICU admission among our cohort, were similar to those published previously, hemodynamic instability and respiratory compromise [9–10]. However, since we included 30-day ICU admissions, additional postoperative conditions were found to contribute to ICU admission as well, including sepsis, anastomotic leak, cardiac and neurologic complications. One benefit of NACT is less disease burden at the time of surgery, meaning less extensive surgical procedures are required to achieve an optimal debulking procedure [12]. Previous studies have associated more radial surgical procedures involving bowel resections and high blood loss with increased likelihood of ICU admission [2]. This was true in our cohort, in which those admitted to the ICU had higher rates of bowel resection, splenectomy and diaphragmatic surgery as well as a higher estimated blood loss. However, with the exception of bowel resection, the rates of these more morbid procedures did not vary among the PDS and NACT patient admitted to the ICU. This suggests that patients who are poor responders to NACT, with diffuse disease at the time of debulking, are at high risk for ICU admission. Among patients admitted to an ICU, there was no difference in length of stay in the ICU between NACT and PDS patients, however PDS patients did have longer overall hospital length of stay. This could be due to higher rates of post-operative complications among PDS patients, though there was not a difference in rates of major postoperative complications between PDS and NACT patients, including readmission, re-operation, pulmonary embolus and 30-day mortality rate. The 30day mortality rate in this cohort was 0.5%, much lower than other reported studies, possibly due to greater use of NACT. Prior studies have shown higher 30-day mortality rates among ICU admitted patient, but in our series, the 3 peri-operative deaths occurred in patients who did not get admitted to the ICU, either due to sudden events or transition of care to hospice. There are several limitations of the current study that must be considered when interpreting the data. Certainly, the retrospective nature of the data could create selection bias. Additionally, as patients were not randomized to PDS vs NACT, and the decision was left up to each individual gynecologic oncologist, there is likely a wide variation in the patients offered NACT. Additionally, patients in this study were not stratified by the number of cycles of neoadjuvant chemotherapy they received. Several prior studies have found serum albumin levels to be significantly associated with ICU admissions, but this data point was not available for inclusion in this analysis [9]. We also do not have data on patients potentially re-admitted within 30 days of surgery to an outside institution. Nor do we propose a reliable way to predict which patients will require ICU care. In conclusion, patients who undergo NACT as opposed to PDS are less likely to require ICU admission. However, those NACT patients requiring ICU care, will likely be those with the most extensive surgical procedures and they will require the same amount of days in the ICU before they are stable to move to the floor. Older age and more extensive metastasis has previously been associated with higher ICU admission rates, but as these patients at highest risk of ICU admission begin to undergo NACT, the rate of post-operative debulking ICU admissions is likely to fall.
ICU admitted
Non-ICU admitted
(n)
% or range
(n)
% or range
P value
12 9 8 0 13.2
28% 21% 18% 0% 5–37
74 12 21 3 6.8
13% b1% 3% b1% 0–24
0.008 b0.001 b0.001 0.810 b0.001
References [1] D. Chalfin, J. Rizzo, Economics of Critical Care, Elsevier, Philadelphia, 2012. [2] T.P. Díaz-Montes, M.L. Zahurak, R.E. Bristow, Predictors of extended intensive care unit resource utilization following surgery for ovarian cancer, Gynecol. Oncol. 107 (3) (2007) 464–468, https://doi.org/10.1016/j.ygyno.2007.07.074.
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028
K. Pepin et al. / Gynecologic Oncology xxx (2017) xxx–xxx [3] S.E. Brooks, J. Ahn, C.D. Mullins, C.R. Baquet, Resources and use of the intensive care unit in patients who undergo surgery for ovarian carcinoma, Cancer 95 (7) (2002) 1457–1462, https://doi.org/10.1002/cncr.10872. [4] A. Melamed, E.M. Hinchcliff, J.T. Clemmer, et al., Trends in the use of neoadjuvant chemotherapy for advanced-stage ovarian cancer: a National Cancer Data Base study, Gynecol. Oncol. 141 (2) (2016) 236–240, https://doi.org/10.1016/j.ygyno. 2016.04.097. [5] I. Vergote, F. Amant, G. Kristensen, T. Ehlen, N.S. Reed, A. Casado, Primary surgery or neoadjuvant chemotherapy followed by interval debulking surgery in advanced ovarian cancer, Eur. J. Cancer 47 (Suppl. 3) (2011)https://doi.org/10.1016/S09598049(11)70152-6. [6] S. Kehoe, J. Hook, M. Nankivell, et al., Primary chemotherapy versus primary surgery for newly diagnosed advanced ovarian cancer (CHORUS): an open-label, randomised, controlled, non-inferiority trial, Lancet 6736 (14) (2015) 1–9, https:// doi.org/10.1016/S0140-6736(14)62223-6. [7] M.C. van Aardt, G. Dreyer, H.F. Pienaar, et al., Unique human papillomavirus-type distribution in South African women with invasive cervical cancer and the effect of human immunodeficiency virus infection, Int. J. Gynecol. Cancer 25 (5) (2015) 919–925, https://doi.org/10.1097/IGC.0000000000000422.
5
[8] G.D. Aletti, S.C. Dowdy, K.C. Podratz, W.A. Cliby, Relationship Among Surgical Complexity, Short-term Morbidity, and Overall Survival in Primary Surgery for Advanced Ovarian Cancer, 2007 1–7 (December) https://doi.org/10.1016/j.ajog.2007.10.495. [9] R. Ruskin, R.R. Urban, A.E. Sherman, et al., Predictors of intensive care unit utilization in gynecologic oncology surgery, Int. J. Gynecol. Cancer 21 (8) (2011) 1336–1342, https://doi.org/10.1097/IGC.0b013e31822d0ed0. [10] C. a Leath, J.E. Kendrick, T.M. Numnum, et al., Outcomes of gynecologic oncology patients admitted to the intensive care unit following surgery: a university teaching hospital experience, Int. J. Gynecol. Cancer 16 (5) (2006) 1766–1769, https://doi. org/10.1111/j.1525-1438.2006.00702.x. [11] M. Amir, MM S, BY K., Surgical intensive care unit care after ovarian cancer surgery: an analysis of indications, Am. J. Obstet. Gynecol. 176 (6) (1997) 1389–1393. [12] J.Y. Hou, M.G. Kelly, H. Yu, et al., Neoadjuvant chemotherapy lessens surgical morbidity in advanced ovarian cancer and leads to improved survival in stage IV disease, Gynecol. Oncol. 105 (1) (2007) 211–217, https://doi.org/10.1016/j.ygyno.2006.11. 025.
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interval debulking surgery Kristen Pepin a,b,⁎, Amy Bregar a, Michelle Davis b, Alexander Melamed a, Emily Hinchcliff c, Allison Gockley b, Neil Horowitz b, Marcela G. del Carmen a a b c
Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114, United States Brigham and Women's Hospital, 75 Francis St, Boston, MA 20115, United States MD Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, United States
H I G H L I G H T S • • • •
Intensive care unit (ICU) admissions are more common in ovarian cancer patients treated with primary debulking (PDS). This was true though patients treated with neoadjuvant chemotherapy (NACT) were older and had more advanced disease. NACT patients admitted to the ICU had comparable surgical complexity at time of cytoreduction as PDS ICU admitted patients. Indications of ICU admission were not different between patients undergoing PDS vs NACT.
a r t i c l e
i n f o
Article history: Received 13 July 2017 Received in revised form 22 September 2017 Accepted 24 September 2017 Available online xxxx Keywords: Intensive care Cytoreduction Neoadjuvant Surgical complexity
a b s t r a c t Objective. Admissions to intensive care units (ICU) are costly, but are necessary for some patients undergoing radical cancer surgery. When compared to primary debulking surgery (PDS), neoadjuvant chemotherapy (NACT) with interval debulking surgery, is associated with less peri-operative morbidity. In this study, we compare rates, indications and lengths of ICU stays among ovarian cancer patients admitted to the ICU within 30 days of cytoreduction, either primary or interval. Methods. A retrospective chart review was performed of patients with stage III-IV ovarian cancer who underwent surgical cytoreduction at two large academic medical centers between 2010 and 2014. Chi square tests, Student t-tests, and Mann-U Whitney tests were used. Results. A total of 635 patients were included in the study. There were 43 ICU admissions, 7% of patients. Compared to NACT, a higher percentage of PDS patients required ICU admission, 9.4% vs 3.9% of patients (P = 0.004). ICU admission indications did not vary between PDS and NACT patients. NACT patients admitted to the ICU had comparable mean surgical complexity scores to those PDS patients admitted to the ICU, 6.2 (95%CI 5.3–7.1) vs 4.5 (95%CI 3.1–6.0) (P = 0.006). Length of ICU admission did not vary between groups, PDS 2.7 days (95%CI 2.3–3.2) vs 3.5 days (95%CI 1.5–5.6) for NACT (P = 0.936). Conclusions. The rate of ICU admissions among patients undergoing PDS is higher than for NACT. Among patients admitted to the ICU, indications for admission, length of stay and surgical complexity were similar between patients treated with NACT and PDS. © 2017 Elsevier Inc. All rights reserved.
1. Introduction Admission to an intensive care unit (ICU) represents a significant cost to the health care system [1]. It has been estimated to account for over half a percent of the United States Gross Domestic Product (GDP) ⁎ Corresponding author at: 89 East Brookline St, Apt 4, Boston, MA 02115, United States. E-mail address: [email protected] (K. Pepin).
[1]. Given the lack of screening for detection of early-stage ovarian cancer, an estimated 85% of patients with epithelial ovarian cancer present with advanced-stage (III/IV) disease and require radical debulking surgery, either upfront or as an interval strategy, which often necessitates ICU-level of care. Post-operative ICU admission rates after ovarian cancer debulking surgery have been reported to be as high as 30% [2]. Predictors of extended ICU stay in these patients include older age, more medical comorbidities and more extensive surgical debulking [3].
https://doi.org/10.1016/j.ygyno.2017.09.028 0090-8258/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028
2
K. Pepin et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Since the publication of two phase III randomized clinical trials, Kehoe et al. and Vergote et al., showing equivalent survival between ovarian cancer patients treated with primary debulking surgery (PDS) and neoadjuvant chemotherapy with interval debulking surgery (NACT), the use of NACT has increased [4–6]. NACT has become a preferred paradigm for the management of patients with unresectable disease upfront or those too medically infirmed to undergo PDS. This is despite the relatively low rates of resection to no residual disease (R0) and lower than previously reported survival, in these trials [5–6]. Though poor candidates for upfront surgery, were historically offered NACT, the before mentioned non-inferiority trials, have likely resulted in a lower the threshold to offer patients NACT. Several studies have reported fewer post-operative complications and lower 30-day mortality among patients treated with NACT vs PDS [5–6]. With the recent shift in treatment of older and more medically complex women with NACT, there have been no accompanying data on how ICU utilization has been affected. In this study, we compare rates, indications and lengths of ICU stays among ovarian cancer patients admitted to the ICU within 30 days of cytoreduction, either primary or interval. 2. Methods We identified all newly diagnosed stage III-IV epithelial ovarian, fallopian tube and primary peritoneal cancer patients who underwent surgical cytoreduction, at two large academic medical centers between 2010 and 2014. A joint institution Institutional Review Board (IRB) approved the retrospective chart review and analysis. Patients were identified using tumor registries and tumor board records. Patients with early stage tumors, germ cell tumors and borderline tumors were not included in the analysis. Patients who were treated exclusively with chemotherapy were excluded from the study, as were patients who underwent debulking surgery at an outside facility. Patients were defined as having had primary debulking surgery if surgery was the first treatment modality the patient underwent, and an attempt at debulking was made at the time of surgery. Patients who received only surgery, and no adjuvant chemotherapy were considered part of the PDS group. The NACT group was defined by any chemotherapy prior to attempt at surgical debulking. In most cases, the planned course was for a debulking procedure following 3 cycles of NACT. The Charlson Comorbidity Index was used to evaluate preoperative comorbid conditions at the time of diagnosis. The index includes predicts 10-year mortality, based on history of myocardial infarction, congestive heart failure, renal disease, diabetes, chronic obstructive pulmonary disease, liver disease, dementia, peripheral vascular disease, peptic ulcer disease, hemiplegia and malignancy. All patients in the study, by definition, had metastatic cancer, earning a Charlson Comorbidity score of 6. For this analysis, their scores were calculated excluding their cancer diagnosis, such that the lowest possible score was zero [7]. Additionally, no condition included in the Comorbidity index should have been significantly improved by use of chemotherapy, thus scores were not recalculated after completion of NACT. Surgical morbidity was characterized by the surgical procedures performed at the time of cytoreduction, including bowel resection, ostomy creation, splenectomy, liver resection and procedures on the diaphragm. All procedures performed were incorporated into a surgical complexity score, to compare the extent of surgical debulking [8]. One point was assigned for hysterectomy/bilateral salpingo-oophorectomy, omentectomy, pelvic lymph node dissection, paraaortic lymph node dissection and small bowel resection. Two points were assigned for large bowel resection, diaphragm stripping/resection, splenectomy and liver resection. Three points were assigned for a rectosigmoid resection with anastomosis. Post-operative morbidity was defined by estimated blood loss (EBL) N 2 l, intensive care unit (ICU) admission, need for re-operation, hospital length of stay (LOS), and readmission within 30-days.
Descriptive statistics were defined with means and medians. Chi square tests, Student t-tests, and Mann-U Whitney tests were used to evaluate categorical and continuous variables. All analysis was performed in SPSS. 3. Results A total of 635 patients met inclusion criteria for the study. There were 43 ICU admissions, just 7% of patients who underwent surgical debulking during this time period. The median age was 64 (range 27–93) and median BMI 25.5 (range 16.1–72.1). Of the included patients, 459 (72%) had stage III disease and 176 (28%) had stage IV disease. Age, race, BMI, diagnosis CA 125 (including CA 125 prior to initiation of chemotherapy for NACT patients), comorbidity index and stage did not vary significantly between patients who were and were not admitted to an ICU post-operatively (Table 1). Those undergoing NACT were older, 64.1 years (range 34–89), than those undergoing PDS, 62 (range 27–93) (P = 0.008) and were more likely to have stage IV disease; 42% of NACT patients vs 14% PDS patients (P = 0.000). Comorbidity index did not vary significantly between the two surgical groups. Rates of NACT rose overtime during the study period (P = 0.001) (Fig. 1). Among the NACT patients, b1% received b 3 cycles of preoperative chemotherapy, 55% received three cycles of preoperative chemotherapy, 28% received four cycles of preoperative chemotherapy and the remainder received N4 cycles of preoperative chemotherapy. Three patients treated with neoadjuvant chemotherapy underwent a laparoscopic debulking procedure, with the rest done via laparotomy. The most common indications for ICU admission were hemodynamic monitoring/pharmacologic blood pressure support (84.1%), respiratory failure/inability to extubate (22.7%), bacteremia/sepsis (18.2%), anastomotic leak (9.1%), cardiac indication (4.6%), neurologic indication (4.5%) and open abdomen (4.5%). Fourteen patients had more than one indication for ICU admission. Patients requiring ICU admission had higher estimated blood loss; 1630 ml (95%CI 1181–2147) vs 503 ml (95%CI 453–540), (P b 0.001). ICU admission indications did not vary between PDS and NACT patients (Fig. 2). Patients admitted to a ICU had significantly higher pre-operative CA − 125 levels; 4125 U/ml (95%CI 702–11,135) vs 545 U/ml (95%CI 435–678), (P b 0.001). Most patients were admitted to the ICU during the initial hospitalization,
Table 1 Patient characteristics. ICU admitted
Age BMI Dx CA 125
Non-ICU admitted
P value
Mean
Range
Mean
Range
64 26.8 6080
39–79 19.0–45.9 4–37,600
63 26.5 1445
26–93 16.1–72.1 22–103,525
Race White Black Hispanic Asian Other/Pt declined Stage 3 4 Histology Serous Other Comorbidity index 0 1 2 3 4 or greater Neoadjuvant chemotherapy
Number
%
Number
%
26 2 0 3 8
61% 5% 0% 7% 27%
470 14 8 18 56
80% 2% 1% 3% 14%
430 162
73% 27%
29 14
67% 33%
387 205
65% 35%
27 16
63% 26%
62 128 166 135 101 12
10% 22% 28% 23% 17% 28%
2 9 15 7 10 294
5% 21% 35% 16% 23% 50%
0.547 0.565 0.267 P value 0.232
0.283
0.952
0.204
0.000
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028
K. Pepin et al. / Gynecologic Oncology xxx (2017) xxx–xxx
3
Table 2 Surgical morbidity. Surgical morbidity
Splenectomy Bowel surgery Liver resection Diaphragm surgery EBL N 2 l Surgical complexity score
Fig. 1. Use of NACT by study year.
but 7% of the ICU admissions occurred during a readmission within 30 days. Compared to NACT, a higher percentage of primary debulking patients required ICU admission, 9.4% vs 3.9% of neoadjuvant chemotherapy patients (P = 0.004). Length of ICU admission did not vary between groups, PDS 2.7 days (95%CI 2.3–3.2) vs 3.5 days (95%CI 1.5–5.6) for NACT (P = 0.936). Total hospital length of stay was longer among PDS vs NACT patients requiring ICU admission, PDS 14.6 days (95%CI 11.3–17.6) vs NACT 9.8 days (95%CI 6.2–13.5), (P = 0.040). Among all patients, median time to first post-operative chemotherapy, though significant varied by just 4 days (PDS = 29 days, 95%CI 28–30 vs NACT = 25 days, 95%CI 24–27, P b 0.001). This was not significantly different for patients who were admitted to an ICU, with a mean time to first postoperative chemotherapy of 27 days (95%CI 21–33). Surgical complexity varied significantly between those admitted and not admitted to the ICU, with splenectomy (35% vs 9%), bowel resection (58%vs 24%), and diaphragmatic surgery (23% vs 10%) being more common in ICU admitted patients (Table 2). Though both groups, ICU admitted and non-ICU admitted had surgical complexity scores ranging 1–11,
ICU admitted
Non-ICU admitted
(n)
% or range
(n)
% or range
15 25 18 10 17 5.7
35% 58% 3% 23% 40% 1–11
52 140 4 59 13 3.4
9% 24% 9% 10% 2% 1–11
P value
b0.001 b0.001 P = 0.054 P = 0.012 b0.001 P = 0.004
those admitted to the ICU had higher mean surgical complexity scores, 5.7 (95%CI 5.0–6.5) vs 3.4 (95%CI 3.2–3.5) (P = 0.006). However, when comparing just those patients admitted to the ICU, rates of splenectomy (39% vs 23%), diaphragmatic surgery (23% vs 23%) and liver resection (10% vs 8%) were similar between those treated with PDS vs NACT (P N 0.05). Patients admitted to the ICU undergoing PDS had higher rates of bowel resection that those undergoing NACT, 71% vs 23% (P = 0.005). Overall surgical complexity scores did not vary significantly between PDS and NACT patients admitted to the ICU, PDS = 6.2 (95%CI 5.3–7.1) vs 4.5 (95%CI 3.1–6.0), (P = 0.600) (Table 3). Patients admitted to the ICU were more likely to have an estimated blood loss N2 l, 40% vs 2% (P b 0.001). Among patients admitted to an ICU, there was not a difference in estimated blood loss between PDS, 1700 cm3 (95%CI 1200–2200) and NACT patients, 1000 cm3 (95%CI 700–2000), (P N 0.50). Patients admitted to an ICU were more likely to be readmitted to the hospital within 30 days (28% vs 13%), undergo reoperation (21% vs b 1%) and to have a postoperative pulmonary embolus (18% vs 3%), as compared with patients not admitted to an ICU (P b 0.01) (Table 4). There were no differences in the rates of these complications between PDS and NACT patients admitted to the ICU; readmission to the hospital within 30 days (29% vs 23%), reoperation (23% vs 15%) and postoperative pulmonary embolus (19% vs 15%), (P N 0.05).
Fig. 2. ICU admission indications by treatment type.
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028
4
K. Pepin et al. / Gynecologic Oncology xxx (2017) xxx–xxx
Table 3 Mean surgical complexity scores. ICU admitted
Non-ICU admitted
PDS
NACT
Mean 95%CI
Mean 95%CI
6.2
5.3–7.1 4.5
P value
PDS
NACT
Mean 95%CI
Mean 95%CI
3.1–6.0 0.600 3.7
3.4–3.9 3.0
P value
2.8–3.2 0.000
4. Conclusions Patients with advanced-stage ovarian cancer often require ICU-level of care. Previously published data have focused on identification of predisposing pre-operative, intraoperative and post-operative factors that increase risk of ICU utilization [9]. However, all literature on ICU admissions among ovarian cancer patients was published during a time when the vast majority of patients were undergoing primary debulking surgery. We analyzed ICU admissions among patients treated between 2010 and 2014, a time during which adoption of NACT increased, following publication of studies showing no difference in overall survival when compared to patient treated by PDS [5–6]. Older age, greater disease burden and more severe medical comorbidities have all been cited as factors predisposing ovarian cancer patients to ICU admission at the time of surgery [3,10]. These may be factors that make patients more likely to be considered for NACT as opposed to PDS. One study showed a 5.9-fold increase in prolonged ICU stay in ovarian cancer patients undergoing debulking surgery over the age of 63 [2]. Patients in our study who received NACT were older and were more likely to have Stage IV disease, though comorbidity index did not vary significantly between groups. The comorbidity index used in this study, while predictive of 10-year mortality, does not capture many conditions that complicate surgical management, including pulmonary embolism, deep venous thrombosis, pleural effusion, morbid obesity, smoking status, immobility, infectious diseases, cardiomyopathy etc. This likely accounts for there being no difference in comorbidity index between PDS and NACT groups, despite NACT generally being intended for more medically complicated patients. The ICU admission rate was significantly lower among patients treated with NACT vs PDS, though NACT patients were on average older and had more advanced disease. This suggests that treating patients with NACT is a way to minimize ICU admissions among the patients most at risk of ICU admission. In our study, 7% of patients undergoing debulking surgery required ICU admission. This rate is on the lower end of what has been previously reported in other studies, with ICU admissions in this patient population reported to range between 6% and 36% [3,9–11]. The rates of ICU admission for patients treated with NACT (3.9%) in our study were significantly lower than those undergoing PDS (9.4%). This finding, may be explained, at least in part, by the fact that prior to the documentation of non-inferiority of NACT when compared to PDS, upfront surgery was performed more frequently and is associated with higher rates of ICU admission. Additionally, most publications on ICU admissions among gynecologic oncology patients focus on only immediate post-operative ICU admissions. In our series, we documented ICU admissions related to Table 4 Post-operative morbidity.
Conflict of interest statement The authors of this manuscript certify that they have no affiliations with or involvement in any organization or entity with a financial interest or non-financial interest in the subject matter discussed in this manuscript.
Post-operative morbidity
Readmit Re-operation DVT/PE Death Hospital length of stay
surgical morbidity, underlying medical comorbidity, and postoperative complications. Thus, we included ICU admissions that occurred within 30 days of surgery. The most common indications for ICU admission among our cohort, were similar to those published previously, hemodynamic instability and respiratory compromise [9–10]. However, since we included 30-day ICU admissions, additional postoperative conditions were found to contribute to ICU admission as well, including sepsis, anastomotic leak, cardiac and neurologic complications. One benefit of NACT is less disease burden at the time of surgery, meaning less extensive surgical procedures are required to achieve an optimal debulking procedure [12]. Previous studies have associated more radial surgical procedures involving bowel resections and high blood loss with increased likelihood of ICU admission [2]. This was true in our cohort, in which those admitted to the ICU had higher rates of bowel resection, splenectomy and diaphragmatic surgery as well as a higher estimated blood loss. However, with the exception of bowel resection, the rates of these more morbid procedures did not vary among the PDS and NACT patient admitted to the ICU. This suggests that patients who are poor responders to NACT, with diffuse disease at the time of debulking, are at high risk for ICU admission. Among patients admitted to an ICU, there was no difference in length of stay in the ICU between NACT and PDS patients, however PDS patients did have longer overall hospital length of stay. This could be due to higher rates of post-operative complications among PDS patients, though there was not a difference in rates of major postoperative complications between PDS and NACT patients, including readmission, re-operation, pulmonary embolus and 30-day mortality rate. The 30day mortality rate in this cohort was 0.5%, much lower than other reported studies, possibly due to greater use of NACT. Prior studies have shown higher 30-day mortality rates among ICU admitted patient, but in our series, the 3 peri-operative deaths occurred in patients who did not get admitted to the ICU, either due to sudden events or transition of care to hospice. There are several limitations of the current study that must be considered when interpreting the data. Certainly, the retrospective nature of the data could create selection bias. Additionally, as patients were not randomized to PDS vs NACT, and the decision was left up to each individual gynecologic oncologist, there is likely a wide variation in the patients offered NACT. Additionally, patients in this study were not stratified by the number of cycles of neoadjuvant chemotherapy they received. Several prior studies have found serum albumin levels to be significantly associated with ICU admissions, but this data point was not available for inclusion in this analysis [9]. We also do not have data on patients potentially re-admitted within 30 days of surgery to an outside institution. Nor do we propose a reliable way to predict which patients will require ICU care. In conclusion, patients who undergo NACT as opposed to PDS are less likely to require ICU admission. However, those NACT patients requiring ICU care, will likely be those with the most extensive surgical procedures and they will require the same amount of days in the ICU before they are stable to move to the floor. Older age and more extensive metastasis has previously been associated with higher ICU admission rates, but as these patients at highest risk of ICU admission begin to undergo NACT, the rate of post-operative debulking ICU admissions is likely to fall.
ICU admitted
Non-ICU admitted
(n)
% or range
(n)
% or range
P value
12 9 8 0 13.2
28% 21% 18% 0% 5–37
74 12 21 3 6.8
13% b1% 3% b1% 0–24
0.008 b0.001 b0.001 0.810 b0.001
References [1] D. Chalfin, J. Rizzo, Economics of Critical Care, Elsevier, Philadelphia, 2012. [2] T.P. Díaz-Montes, M.L. Zahurak, R.E. Bristow, Predictors of extended intensive care unit resource utilization following surgery for ovarian cancer, Gynecol. Oncol. 107 (3) (2007) 464–468, https://doi.org/10.1016/j.ygyno.2007.07.074.
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028
K. Pepin et al. / Gynecologic Oncology xxx (2017) xxx–xxx [3] S.E. Brooks, J. Ahn, C.D. Mullins, C.R. Baquet, Resources and use of the intensive care unit in patients who undergo surgery for ovarian carcinoma, Cancer 95 (7) (2002) 1457–1462, https://doi.org/10.1002/cncr.10872. [4] A. Melamed, E.M. Hinchcliff, J.T. Clemmer, et al., Trends in the use of neoadjuvant chemotherapy for advanced-stage ovarian cancer: a National Cancer Data Base study, Gynecol. Oncol. 141 (2) (2016) 236–240, https://doi.org/10.1016/j.ygyno. 2016.04.097. [5] I. Vergote, F. Amant, G. Kristensen, T. Ehlen, N.S. Reed, A. Casado, Primary surgery or neoadjuvant chemotherapy followed by interval debulking surgery in advanced ovarian cancer, Eur. J. Cancer 47 (Suppl. 3) (2011)https://doi.org/10.1016/S09598049(11)70152-6. [6] S. Kehoe, J. Hook, M. Nankivell, et al., Primary chemotherapy versus primary surgery for newly diagnosed advanced ovarian cancer (CHORUS): an open-label, randomised, controlled, non-inferiority trial, Lancet 6736 (14) (2015) 1–9, https:// doi.org/10.1016/S0140-6736(14)62223-6. [7] M.C. van Aardt, G. Dreyer, H.F. Pienaar, et al., Unique human papillomavirus-type distribution in South African women with invasive cervical cancer and the effect of human immunodeficiency virus infection, Int. J. Gynecol. Cancer 25 (5) (2015) 919–925, https://doi.org/10.1097/IGC.0000000000000422.
5
[8] G.D. Aletti, S.C. Dowdy, K.C. Podratz, W.A. Cliby, Relationship Among Surgical Complexity, Short-term Morbidity, and Overall Survival in Primary Surgery for Advanced Ovarian Cancer, 2007 1–7 (December) https://doi.org/10.1016/j.ajog.2007.10.495. [9] R. Ruskin, R.R. Urban, A.E. Sherman, et al., Predictors of intensive care unit utilization in gynecologic oncology surgery, Int. J. Gynecol. Cancer 21 (8) (2011) 1336–1342, https://doi.org/10.1097/IGC.0b013e31822d0ed0. [10] C. a Leath, J.E. Kendrick, T.M. Numnum, et al., Outcomes of gynecologic oncology patients admitted to the intensive care unit following surgery: a university teaching hospital experience, Int. J. Gynecol. Cancer 16 (5) (2006) 1766–1769, https://doi. org/10.1111/j.1525-1438.2006.00702.x. [11] M. Amir, MM S, BY K., Surgical intensive care unit care after ovarian cancer surgery: an analysis of indications, Am. J. Obstet. Gynecol. 176 (6) (1997) 1389–1393. [12] J.Y. Hou, M.G. Kelly, H. Yu, et al., Neoadjuvant chemotherapy lessens surgical morbidity in advanced ovarian cancer and leads to improved survival in stage IV disease, Gynecol. Oncol. 105 (1) (2007) 211–217, https://doi.org/10.1016/j.ygyno.2006.11. 025.
Please cite this article as: K. Pepin, et al., Intensive care admissions among ovarian cancer patients treated with primary debulking surgery and neoadjuvant chemotherapy-interv..., Gynecol Oncol (2017), https://doi.org/10.1016/j.ygyno.2017.09.028