Archives of Gynecology and Obstetrics

, Volume 284, Issue 2, pp 445–451

Malnutrition as a predictor of poor postoperative outcomes in gynecologic cancer patients

Authors

    • Department of Obstetrics and Gynecology, Miller School of Medicine, Jackson Memorial HospitalUniversity of Miami
  • Kathleen F. Brookfield
    • Department of Obstetrics and Gynecology, Miller School of Medicine, Jackson Memorial HospitalUniversity of Miami
  • Samer I. Schuman
    • Division of Gynecologic Oncology, Department of Obstetrics and GynecologyMiller School of Medicine, Jackson Memorial Hospital
  • Joseph A. LucciIII
    • Division of Gynecologic Oncology, Department of Obstetrics and GynecologyMiller School of Medicine, Jackson Memorial Hospital
Gynecologic Oncology

DOI: 10.1007/s00404-010-1659-y

Cite this article as:
Kathiresan, A.S.Q., Brookfield, K.F., Schuman, S.I. et al. Arch Gynecol Obstet (2011) 284: 445. doi:10.1007/s00404-010-1659-y

Abstract

Purpose

Poor nutritional status has been associated with increased postoperative morbidity and mortality in surgical patients. The purpose of this study is to evaluate if decreased nutritional parameters correlate with increased postoperative complications regardless of other risk factors in the gynecologic cancer patient.

Methods

A retrospective chart review was performed among women who underwent surgical management for gynecologic malignancies from October 2006 to June 2008. Variables included age, race, medical comorbidities, cancer type/stage, preoperative albumin, absolute lymphocyte count (ALC), and body mass index (BMI), estimated blood loss (EBL), intraoperative blood transfusion (BT), intraoperative or postoperative complications, intensive care unit (ICU) admissions, hospital readmissions, reoperations, and cancer recurrence.

Results

Three hundred gynecologic oncology patients with preoperative nutritional parameters were included in the study. Decreased albumin was significantly associated with more postoperative complications (p < 0.001), hospital readmissions (p = 0.01), reoperations (p = 0.03), ICU admissions (p < 0.001), and cancer recurrence (p < 0.001). Decreased ALC and BMI preoperatively was also significantly associated with higher incidence of cancer recurrence (p = 0.01, p = 0.01). Surgical cases involving increased EBL (p = 0.01, p < 0.001) and more BT (p < 0.001, p < 0.001) had significantly more postoperative complications and more ICU admissions. Multivariable logistic regression found preoperative albumin to be an independent predictor of increased postoperative complications.

Conclusions

Decreased albumin is significantly associated with more postoperative complications, hospital readmissions, reoperations, ICU admissions, and cancer recurrence. This nutritional parameter is an important predictor of postoperative morbidity and mortality. Thus, it is important to assess nutritional status preoperatively and offer nutritional support or alternate treatment options if necessary.

Keywords

MalnutritionAlbuminPostoperative complicationsGynecology oncology

Introduction

Malnutrition frequently coexists with chronic disease; among gynecologic cancer patients, the prevalence of malnutrition is approximately 20% at time of diagnosis [1]. It has been suggested that up to 20% of patients with cancer die from the effects of malnutrition than from the malignancy itself [2]. In the setting of excessive metabolic demands caused by chronic illness, the body’s ability to conserve protein is compromised. If the increased needs are not met from dietary or therapeutic sources, visceral protein stores are depleted leading to gastrointestinal malabsorption, impaired immunologic response, and impaired production of plasma proteins in the liver [3, 4]. As such, the nutritional status of gynecologic cancer patients can be evaluated using various anthropometric measures (i.e., weight loss, body mass index (BMI), triceps skinfold thickness, and arm circumference), immunologic measurements (i.e., absolute lymphocyte count and delayed cutaneous hypersensitivity response to skin test antigens) and serum protein markers (i.e., albumin, prealbumin, transferrin, and retinol-binding protein). The most commonly used laboratory markers for nutritional status are serum protein markers, and some are more appropriate for short-term or long-term assessment of nutritional status, depending upon the half-life (albumin half-life = 20 days, transferrin half-life = 8–10 days, prealbumin half-life = 2–3 days, retinal-binding protein half-life = 12 h) [5].

Adequate nutrition status affects gynecologic oncology patients’ survival and quality of life. More specifically, it can improve a patient’s ability to tolerate oncological therapies including surgery, chemotherapy, and radiation [5]. In the surgical patient, the association of malnutrition with poor postoperative outcomes has been well established in fields including gastrointestinal, cardiovascular, orthopedic, neurosurgical, and vascular surgery; however, the literature specific to the gynecologic oncologic patient is limited [611].

As previous studies have been based on selected types of gynecologic cancers and small sample sizes, they have failed to separate the predictive ability of nutritional status to determine adverse surgical outcomes from other risk factors. The purpose of this study is to evaluate nutritional parameters preoperatively and determine if malnutrition correlates with increased postoperative morbidity and mortality regardless of other risk factors. To our knowledge, this is the largest study reported on this topic, and the first to report on morbidities including hospital readmissions, intensive care unit admissions, reoperations, and cancer recurrence in this patient population.

Materials and methods

After receiving approval from our IRB committee, a retrospective chart review was performed from October 2006 to June 2008 reviewing gynecologic oncology patients who had undergone surgery at our institution, University of Miami/Jackson Memorial Hospital in Miami, Florida. Inclusion criteria included women with a diagnosis of cancer on pathology with preoperative nutritional parameters including albumin, BMI, and absolute lymphocyte count (ALC). Each patient’s disease was evaluated and staged according to the International Federation of Gynecology and Obstetrics (FIGO) classification.

Variables collected included age, race, cancer type and stage, number of comorbidities, preoperative ALC, preoperative albumin, preoperative BMI, estimated blood loss (EBL), intraoperative blood transfusion (BT), intraoperative complications, postoperative complications, hospital readmissions, intensive care unit (ICU) admissions, reoperations, and cancer recurrence. Preoperative ALC and albumin were further divided into three discreet intervals, namely, below normal range, normal range, and above normal range, based on our institutional definition of normal. BMI was also divided into four subgroups, namely, underweight (BMI < 18.49 kg/m2), normal (BMI 18.5–24.9 kg/m2), overweight (BMI 25–29.9 kg/m2), and obese (BMI > 30 kg/m2).

Albumin was included if the value was obtained within 20 days of surgery as this would be an accurate reflection of nutrition status prior to surgery. Our institutional normal range is 3.9–5.0 g/dL for albumin and 1.0 to 3.3 (×103/mL) for ALC. Intraoperative complications encountered included unintended bladder or bowel injury. All postoperative complications, hospital readmissions, ICU admissions, and reoperations were recorded if they took place within 30 days of surgery. Postoperative complications included pulmonary embolus, deep vein thrombosis, pneumonia, would infection or dehiscence, coagulopathies, anastomotic leaks or fistulas, respiratory failure, sepsis, or death <30 days after surgery. The extent of the surgery was quantified by the duration of the surgical case. A surgery was considered major if it was more than 200 min in duration.

Because ovarian cancer often present at more advanced stages needing extensive debulking and bowel resections, the morbidity associated with this cancer type is greater. Therefore, our data was further subdivided into patients with ovarian cancer versus other gynecologic cancers and patients with bowel resection versus no bowel resection.

The data was then analyzed using SPSS (version 17.0) statistical software. Continuous variables including age, number of comorbidities, EBL, ALC, albumin level, and BMI were compared between patients experiencing complicated postoperative courses and those not experiencing complications using the ANOVA test and Student’s t test. Categorical variables, including race, ALC, albumin, and BMI divided into discreet intervals, need for intraoperative BT, incidence of intraoperative complications, location of cancer (ovarian vs. other gynecologic cancer), and occurrence of intestinal resection and their association with postoperative complications were analyzed using the Chi-square test or Fisher’s exact test, where appropriate. Multivariable logistic regression was performed to analyze which nutritional parameters were independent predictors of increased postoperative complications. A p value of <0.05 was considered statistically significant.

Results

Three hundred gynecologic oncology patients with preoperative nutritional parameters underwent surgical management. The distribution of their primary cancer site was 71 ovarian (23.7%), 106 cervical (35.3%), 86 endometrial (28.7%), and 10 vulvar (3.3%). Patient characteristics including age and race were similar across all groups, with the exception of Caucasian patients being significantly more likely to be admitted to the ICU and Hispanic patients being significantly more likely to develop cancer recurrence (Table 1). The Hispanic population made up the majority of the sample size at 52%, followed by Caucasians at 28%, and African Americans at 20%. A lower mean number of comorbidities was significantly associated with fewer postoperative complications (1.2 vs. 1.8 vs. 1.6; p = 0.02; Table 1).
Table 1

Demographic characteristics, intraoperative findings, and nutritional parameters of the study group according to postoperative complications, readmissions, reoperations, ICU admissions, and cancer recurrence

 

Entire cohort

Number of postoperative complications

Readmissions

Reoperations

ICU admissions

Cancer recurrence

n

0

1

≥2

p

No

Yes

p

No

Yes

p

No

Yes

p

No

Yes

p

Age

300

54.4

55.9

52.4

0.41

54.7

51.2

0.11

54.0

53.8

0.95

54.3

52.4

0.50

54.3

53.1

0.53

Mean number of comorbidities

300

1.2

1.8

1.6

0.02

1.4

1.3

0.74

1.4

1.3

0.86

1.3

2.2

0.004

1.4

1.2

0.38

Mean EBL

300

377.8

593.7

680.5

0.01

485.6

446.3

0.73

490.6

306.9

0.29

415.2

1212.4

<0.001

459.7

566.4

0.31

Mean ALC

300

2.3

2.0

1.7

0.32

2.2

1.9

0.49

2.1

1.6

0.35

2.2

1.3

0.12

2.3

1.4

0.01

Mean Albumin

300

4.1

3.7

3.4

<0.001

3.9

3.5

0.01

3.8

3.4

0.03

3.9

3.0

<0.001

4.0

3.3

<0.001

Mean BMI

300

30.6

32.7

27.9

0.07

31.0

28.1

0.10

30.7

27.8

0.24

30.5

30.4

0.96

31.2

27.0

0.01

% of total

 Race

  Caucasian

83

31.1

25.0

24.0

0.8

30.4

17.4

0.19

28.1

33.3

0.5

26.6

48.0

0.03

24.9

40.7

0.04

  AA

60

20.3

23.2

20.0

19.8

21.7

19.5

27.8

19.9

24.0

21.9

13.6

  Hispanic

153

48.6

51.8

56.0

49.8

60.9

52.4

38.9

53.6

28.0

53.2

45.8

 ALC groups

  <1.0

41

11.0

17.0

31.8

0.02

16.2

17.5

0.98

15.5

35.3

0.07

14.1

40.9

0.01

8.5

45.3

<0.001

  1.0–3.3

193

82.1

71.7

63.6

76.2

75.0

77.0

52.9

78.0

54.5

82.6

50.9

  >3.3

20

6.9

11.3

4.5

7.6

7.5

7.5

11.8

7.9

4.5

9.0

3.8

 Albumin groups

  <3.89

76

30.5

66.7

57.9

0.002

41.7

62.2

0.08

44.8

62.5

0.39

42.0

80.0

0.006

33.9

73.5

<0.001

  3.9–5.0

89

68.3

33.3

42.1

57.5

37.8

54.5

37.5

57.3

20.0

64.4

26.5

  >5.1

2

1.2

0.0

0.0

0.8

0.0

0.7

0.0

0.7

0.0

1.7

0.0

 BMI groups

  <18.49

5

1.1

0.0

12.0

0.29

0.8

17.4

0.001

3.0

9.1

0.67

3.6

0.0

0.58

1.6

12.5

0.01

  18.5–24.9

32

25.0

10.3

24.0

22.2

17.4

22.0

27.3

20.4

36.4

18.9

33.3

  25–29.9

41

27.3

24.1

28.0

26.2

30.4

25.8

27.3

27.7

18.2

28.3

20.8

  >30

73

46.6

65.5

36.0

50.8

34.8

49.2

36.4

48.2

45.5

51.2

33.3

 Intraop BT

  No

242

88.6

75.4

66.0

<0.001

81.7

82.6

1.0

82.3

72.2

0.34

85.0

48.0

<0.001

83.1

76.3

0.26

  Yes

54

11.4

24.6

34.0

18.3

17.4

17.7

27.8

15.0

52.0

16.9

23.7

 Intraop Compl

  No

284

97.2

89.5

96.0

0.054

95.1

97.8

0.70

95.9

88.9

0.19

96.3

88.0

0.09

96.2

93.2

0.30

  Yes

13

2.8

10.5

4.0

4.9

2.2

4.1

11.1

3.7

12.0

3.8

6.8

 Ovarian cancer

  No

224

77.7

64.3

80.0

0.09

74.3

82.6

0.27

75.8

72.2

0.78

77.1

62.5

0.13

79.7

61.0

0.004

  Yes

71

22.3

35.7

20.0

25.7

17.4

24.2

27.8

22.9

37.5

20.3

39.0

 Intestinal resection

  No

260

93.2

78.9

76.0

0.001

88.3

80.4

0.15

89.2

50.0

<0.001

88.4

72.0

0.03

95.0

55.9

<0.001

  Yes

38

6.8

21.1

24.0

11.7

19.6

10.8

50.0

11.6

28.0

5.0

44.1

EBL estimated blood loss; ALC absolute lymphocyte count; BMI body mass index; AA African American

In our study, 177 (62.3%) experienced no postoperative complication, 57 (20.0%) experienced one postoperative complication, and 50 (17.6%) experienced two or more postoperative complications. Those patients who had two or more postoperative complications had significantly more intraoperative blood transfusions (11.4 vs. 24.6 vs. 34.0%; p < 0.001) and higher EBL during surgery (377.8 vs. 593.7 vs. 680.5 mL; p = 0.01; Table 1). Higher EBL (415.2 vs. 1212.4 mL; p < 0.001), more intraoperative blood transfusions (15 vs. 52%; p < 0.001), and increased number of comorbidities (1.3 vs. 2.2; p = 0.004) correlated with significantly more ICU admissions (Table 1).

Decreased albumin was significantly associated with more postoperative complications (4.1 vs. 3.7 vs. 3.4 g/dL; p < 0.001), more hospital readmissions (3.9 vs. 3.5 g/dL; p = 0.01), more reoperations (3.8 vs. 3.4 g/dL; p = 0.03), more ICU admissions (3.9 vs. 3.0 g/dL; p < 0.001), and more cancer recurrence (4.0 vs. 3.3 g/dL; p < 0.001; Table 1). In addition, decreased ALC (2.3 × 103/mL vs. 1.4 × 103/mL; p = 0.01) and BMI (31.2 vs. 27.0 kg/m2; p = 0.01) were also significantly correlated with higher incidence of cancer recurrence (Table 1).

When divided into intervals based on institutional normal range, lower ALC (<1.0 × 103/mL) and lower albumin levels (<3.89 g/dL) were significantly associated with more postoperative complications (11 vs. 17 vs. 31.8%; p = 0.02; 0 vs. 42.1 vs. 57.9%; p = 0.002), ICU admissions (14.1 vs. 40.9%; p = 0.01; 42 vs. 80%; p = 0.006), and cancer recurrence (8.5 vs. 45.3%; p < 0.001; 33.9 vs. 73.5%; p < 0.001; Table 1). In addition, underweight subjects with BMI < 18.49 kg/m2 significantly correlated with more hospital readmissions (0.8 vs. 17.4%; p = 0.001) and more cancer recurrence (1.6 vs. 12.5%; p = 0.01; Table 1).

When the patients were subdivided based on site of cancer, patients with ovarian cancer had significantly lower albumin levels (3.9 ± 0.8 vs. 3.6 ± 0.8; p = 0.02) and more intraoperative blood transfusions (14.9 vs. 28.2%; p = 0.02), intraoperative complications (2.2 vs. 11.3%; p = 0.004), and cancer recurrence (20.3 vs. 39.0%; p = 0.004). When albumin levels were divided into discrete intervals, a greater proportion of ovarian cancer patients had albumin levels below the institutional normal range (<3.89 g/dL), which nearly reached statistical significance (39.8 vs. 60.5%; p = 0.054; Tables 1, 2).
Table 2

Demographic characteristics, intraoperative findings, and nutritional parameters of the study group according to type of cancer (ovarian cancer vs. other gynecologic cancers) and intestinal resection

 

Cancer type

Intestinal resection

Other gyn ca

Ovarian ca

p

No

Yes

p

Age

53.7 ± 13.1

55.1 ± 13.6

0.42

54.2 ± 13.4

53.7 ± 12.7

0.86

Mean number of comorbidities

1.3 ± 1.5

1.5 ± 1.6

0.43

1.4 ± 1.5

1.3 ± 1.6

0.83

Mean EBL

437.0 ± 633.5

542.7 ± 610.6

0.22

426.4 ± 585.1

835.3 ± 1186.6

0.001

Mean ALC

2.2 ± 2.8

1.8 ± 0.7

0.26

2.2 ± 2.6

1.4 ± 0.8

0.06

Mean albumin

3.9 ± 0.8

3.6 ± 0.8

0.02

3.9 ± 0.7

3.3 ± 0.8

<0.001

Mean BMI

30.7 ± 7.8

29.9 ± 7.0

0.64

30.8 ± 7.6

28.5 ± 7.8

0.29

% of total

 Race

  Caucasian

24.6

36.6

0.08

27.1

34.2

0.60

  AA

23.2

14.1

 

20.2

21.1

  Hispanic

52.2

49.3

 

52.7

44.7

 ALC groups

  <1.0

17.1

13.3

0.45

12.7

38.2

0.001

  1.0–3.3

74.1

81.7

 

78.6

58.8

  >3.3

8.8

5.0

 

8.6

2.9

 Albumin groups

  <3.89

39.8

60.5

0.054

38.5

75.0

0.001

  3.9–5.0

58.5

39.5

 

60.0

25.0

  >5.1

1.6

0

 

1.5

0

 BMI groups

  <18.49

4.2

0

0.71

2.9

7.1

0.71

  18.5–24.9

20.0

24.1

 

20.4

28.6

  25–29.9

27.5

27.6

 

27.7

21.4

  >30

48.3

48.3

 

48.9

42.9

 Intraop BT

  No

85.1

71.8

0.02

84.9

60.5

0.001

  Yes

14.9

28.2

 

15.1

39.5

 Intraop Compl

  No

97.8

88.7

0.004

96.1

92.1

0.22

  Yes

2.2

11.3

 

3.9

7.9

EBL estimated blood loss; ALC absolute lymphocyte count; BMI body mass index; AA African American

Cancer patients who underwent intestinal resection experienced significantly greater EBL (426.4 ± 585.1 vs. 835.3 ± 1186.6; p = 0.001), more intraoperative blood transfusions (15.1 vs. 39.5%; p = 0.001), more postoperative complications (6.8% vs. 21.1 vs. 24.0; p = 0.001), more reoperations (10.8 vs. 50.0%; p < 0.001), more ICU admissions (11.6 vs. 28.0%; p = 0.03), and more cancer recurrence (5.0 vs. 44.1%; p < 0.001). Lower albumin levels, both as a continuous variable (3.9 ± 0.7 vs. 3.3 ± 0.8; p < 0.001) and when divided into discrete intervals (<3.89 g/dL: 38.5 vs. 75.0%; p = 0.001), and lower ALC (<1.0 × 103/mL: 12.7 vs. 38.2%; p = 0.001) were significantly associated with more intestinal resections (Tables 1, 2). A multivariable logistic regression model demonstrated albumin was an independent predictor of increased postoperative complications (Table 3). When albumin was below normal range (<3.89 g/dL), patients were 3.44 times more likely to develop a postoperative complication.
Table 3

Logistic regression analysis to assess the relationship between albumin and postoperative complications

Variable

Coefficient

SE

Odds ratio (95% CI)

p

Age

−0.01

0.02

0.99 (0.95, 1.03)

0.64

Race

 Caucasian

  

1.00

 

 AA

0.38

0.80

1.46 (0.31, 6.97)

0.63

 Hispanic

0.50

0.61

1.66 (0.5, 5.49)

0.41

Comorbidities

0.14

0.20

1.15 (0.78, 1.70)

0.49

EBL

0.00

0.00

1.00 (1.00, 1.00)

0.53

Albumin

 <3.89 g/dL

  

1.00

 

 >3.9 g/dL

−1.24

0.51

0.29 (0.11, 0.78)

0.01

Stage

 1

  

1.00

 

 2

0.51

0.63

1.67 (0.49, 5.71)

0.42

 3

1.34

0.65

3.80 (1.07, 13.52)

0.04

 4

1.54

1.14

4.68 (0.50, 44.03)

0.18

Surgical type

 <199 min

  

1.00

 

 >200 min

−0.01

0.49

0.99 (0.38, 2.58)

0.98

Site of cancer

 Other gyn cancer

  

1.00

 

 Ovarian cancer

−0.76

0.64

0.47 (0.13, 1.65)

0.24

Intestinal resection

 No

  

1.00

 

 Yes

−0.07

0.74

0.93 (0.22, 3.95)

0.93

AA African American; EBL estimated blood loss

Discussion

Few studies have investigated preoperative nutritional status as a predictor of postoperative complications in gynecologic cancers. Two previous studies evaluated this topic in isolated types of gynecologic malignancies. In a study by Geisler et al., prealbumin was evaluated as a criterion to determine whether cytoreductive surgery ought to be performed for ovarian cancer and correlated prealbumin levels with occurrence of postoperative complications. They concluded that postoperative complications were indeed more common in the group with the lowest prealbumin [12]. Another study by Alphs et al. investigated albumin levels as a preoperative predictor of surgical outcome and survival among elderly women with ovarian or primary peritoneal cancer. They found decreased albumin levels were associated with significantly higher risk of mortality and suboptimal cytoreduction [13]. Even fewer studies have investigated nutritional parameters correlated to postoperative complications in various types and stages of gynecologic cancers. In a study by Santoso et al., length of hospital stay was used as an indirect indicator of hospital complication rates. They found malnutrition to be significantly correlated to longer hospital stay independent of age, extent of disease or primary tumor site. However, given the small sample size of 67 patients in this study, they were unable to evaluate for other major morbidities including hospital readmissions, reoperations, ICU admissions, and cancer recurrence rates [14].

In our study, decreased albumin was significantly associated with more postoperative complications, hospital readmissions, reoperations, ICU admissions, and cancer recurrence. In addition, albumin was found to be independent predictor of increased postoperative complications. This association remained significant after controlling for other variables including age, race, comorbidities, cancer stage, EBL, extent of surgery determined by operative time, location of cancer, and whether the patient underwent intestinal resection. When divided into discrete intervals based on our institutional normal range, a cutoff value of albumin < 3.89 g/dL significantly correlated with more postoperative complications, ICU admissions, and cancer recurrence. Our study also found low ALC and BMI were significantly correlated with predicting cancer recurrence. We were unable to include transferrin, retinol-binding protein, or prealbumin as a nutritional parameter in our study as these laboratory values were not measured routinely prior to surgery. Regardless, albumin may serve as a better reflection of nutritional status compared to other protein markers with shorter half-lives as their values may fluctuate prior to surgery if the patient has been fasting for an extended period. Based on our results, albumin appears to have a strong predictive value in determining postoperative morbidity. This finding correlates well with prior studies that have suggested that albumin is a good substitute for other more complex, time-consuming tools to assess for malnutrition in gynecologic cancer patients [14]. Thus, our findings support the practice of routinely assessing albumin levels preoperatively.

Intuitively, given the results of our study, nutritional supplementation plans for malnourished patients preoperatively seems warranted to decrease the likelihood of postoperative morbidity and improve long-term patient outcomes. Yet, no studies have evaluated the value of preoperative nutritional supplementation on surgical outcomes in the gynecologic cancer patient; and early literature on the surgical patient have suggested that preoperative total parenteral nutrition (TPN) in severely malnourished patients may be associated with lower rates of postoperative complications [1520]. However, more recent literature and meta-analyses have shown that TPN has no effect on mortality. Furthermore, TPN may reduce complication rates especially in malnourished patients, but the beneficial effects of TPN are seen in those studies done prior to 1988 with low methodologic quality. Therefore, given the increased cost and complications associated with TPN, further studies are needed to confirm the beneficial effects of TPN on postoperative outcomes in the surgical and the gynecologic cancer patient prior to recommending nutritional supplementation preoperatively [16, 1820].

Weaknesses of the study include the retrospective nature and the limitation to the medical records solely at our institution. If patients sought medical care for their postoperative complications at other facilities, these data were not available for our study. Additionally, a comparison of albumin to other nutrition parameters including prealbumin, transferrin, and retinol-binding protein would have been useful in determining which preoperative assessment is superior in predicting postoperative morbidity. Future randomized controlled trials assessing this would be beneficial to optimally determine what impact nutrition status has on patient outcomes and as to which serum marker best predicts nutritional status and postoperative morbidity. Our study found that malnutrition reflected by low albumin levels is associated with significantly higher postoperative morbidity in gynecologic cancer patients, and the gynecologic oncologist should consider routine assessment of patient nutritional status prior to surgical intervention and possibly early nutritional support or alternatives treatment options if warranted.

Conflict of interest

None.

Copyright information

© Springer-Verlag 2010