Gynecologic Oncology 133 (2014) 298–303

Contents lists available at ScienceDirect

Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno

The influence of comorbidity on mortality in ovarian cancer patients K. Stålberg a,⁎, T. Svensson b, S. Lönn c,d, H. Kieler b a

Department of Women's and Children's Health, Uppsala University, 75185 Uppsala, Sweden Centre for Pharmacoepidemiology, Department of Medicine, Karolinska Institutet, 17177 Stockholm, Sweden c Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, 17177 Stockholm, Sweden d Research and Development, Region Halland, 30185 Halmstad, Sweden b

H I G H L I G H T S • The influence of common comorbidity diagnoses on mortality was evaluated in more than 11000 women with ovarian cancer. • Thromboembolism, hematologic complications and infections had a pronounced effect on mortality rates. • The impact of comorbidity was most apparent among those with less aggressive tumours and younger age.

a r t i c l e

i n f o

Article history: Received 14 December 2013 Accepted 15 February 2014 Available online 25 February 2014 Keywords: Ovarian cancer Comorbidity Mortality Epidemiology

a b s t r a c t Background. Ovarian cancer is a severe disease with a peak incidence in the older age groups where concurrent morbidity is common and could potentially influence mortality rates. Objectives. The aim was to study the influence of common comorbidity diagnoses on mortality in ovarian cancer patients. Methods. The study population was patients with ovarian cancer in Sweden 1993–2006 (n = 11.139) identified in the national Cancer Register. Comorbidity data was obtained from the Patient Register and mortality from Cause of Death Register. Mortality was analyzed with Cox' proportional hazards models and subgroup analyses were performed by age and tumor histology. Results. Almost all of the assessed comorbidities increased mortality in ovarian cancer patients. Thromboembolism was the most hazardous comorbidity (HR = 1.95, b1 year after cancer diagnosis and HR = 7.83, 1–5 years after cancer diagnosis) followed by hematologic complications (HR = 1.84 and 7.11 respectively) and infectious disease (HR = 1.48 and 5.28 respectively). The occurrence of diabetes mellitus and hypertension had less impact on mortality. Conclusion. Thromboembolism, hematologic complications and infections had a pronounced effect on mortality rates in women with ovarian cancer. The impact of comorbidity was mainly apparent among those with a more prosperous prognosis, such as longer time since cancer diagnosis, less aggressive tumors and younger age. © 2014 Elsevier Inc. All rights reserved.

Introduction Ovarian cancer has the highest mortality rate among all gynecological malignancies with around 40% surviving five years [12,23]. The incidence, which is 10.4 per 100,000 person years, is increasing by age and is approximately five times higher for women over 65 years of age [7]. The relation to higher age makes the presence of other diseases an important factor affecting cancer morbidity [22] and mortality. Both chronic comorbidities and acute conditions influence the treatment and prognosis of ovarian cancer [5,6,8]. In addition

age and comorbidity influence the choice of treatment and prognosis [14–17]. The aim of this study was to estimate the effect of comorbidity on mortality in patients with ovarian cancer in relation to time of cancer diagnosis. We used information from Swedish health registers and assessed the influence of age at diagnosis and histologic tumor type. Methods The Cancer and Comorbidity Database

⁎ Corresponding author. E-mail address: [email protected] (K. Stålberg).

http://dx.doi.org/10.1016/j.ygyno.2014.02.024 0090-8258/© 2014 Elsevier Inc. All rights reserved.

The Cancer and Comorbidity Database (CaCom) is a compilation of Swedish national health registries as described previously [22]

K. Stålberg et al. / Gynecologic Oncology 133 (2014) 298–303

and contains all cases of cancer registered in the Swedish National Cancer Register (NCR) between 1992 and 2006. The database includes approximately 1.5 million control individuals randomly sampled from the Register of the Total Population (RTP) held by Statistics Sweden, and matched to have the same distribution with respect to age, sex and calendar year as all cases of cancer in the database. These individuals (cancer cases and population controls) are linked to the nationwide Swedish Cause of Death Register to obtain date and cause of death for deceased individuals, and with the Swedish Patient Register to obtain in-patient discharge data for all hospitalizations in these individuals occurring from 1987 up to 2006. In this study we used the database to explore predefined comorbidities among patients with ovarian cancer as well as in a matched control group. Study population The study population was defined as all patients in the CaCom database diagnosed with ovarian cancer from January 1st 1993 up to November 30th 2006. The patients should be alive at 30 days after diagnosis in order to exclude those “fatal at diagnosis” and restrict the study to ovarian cancer patients potentially amenable to treatment in routine medical care. Patients with a pathology diagnosis (SNOMED, Systematized Nomenclature of Medicine, classification) of borderline tumor or tumors considered as benign (i.e. thecoma, cystadenoma) were excluded from the analyses. Pathological diagnoses were grouped into six categories: high risk epithelial including serous and endometroid adenocarcinomas, anaplastic, small cell and poorly differentiated carcinomas; medium risk epithelial including mucinous adenocarcinomas, clear cell carcinomas, and mesonefroid cancers; germ cell tumors, stroma cell tumors and sarcomas, which were classified according to standard definitions. The category other tumors included varying histology such as poorly defined tumors, squamous cell cancers and neuroendocrine tumors. Control group To each cancer case, five controls matched by year of birth, were randomly selected among the 0.75 million control women in the CaCom database. Each control was assigned the same date for start of followup as her matched case; this date was the date of diagnosis for ovarian cancer plus an additional 30 days, and defined as the index date. The controls were not allowed to have an ovarian cancer diagnosed before the corresponding matched case date of diagnosis, and were censored on the relevant date if they contracted an ovarian cancer diagnosis during the follow-up. Comorbidities In CaCom we identified comorbidities by diagnostic codes according to the ICD-classification, ninth and tenth revision (ICD-9 and ICD-10). We selected 6 major disease groups; hematologic complications, thromboembolic event, cardiovascular disease (excluding acute cardiac infarction), infectious disease, diabetes mellitus and hypertension [22]. Comorbidity was registered between one year before ovarian cancer diagnosis and five year after diagnosis. Sources of data The Swedish Cancer Register was founded in 1958 and contains information about clinical and histological diagnoses and date and place of living at diagnosis. It is updated annually and reported valid information on more than 97% of all patients with cancer [2]. The Cancer Register converts all diagnoses recorded by the current ICD and ICDO version into ICD 7 in order to facilitate comparisons over time.

299

The Swedish Patient Register includes data on dates of each hospital admission, main discharge diagnosis and secondary diagnoses by ICD codes. The routines of recording and forwarding diagnoses are standardized across Sweden. The Cause of Death Register contains data on dates and causes of death for Swedish residents since 1961. The coverage is more than 99.5%.

Statistical analyses Relative 5-year mortality rates were calculated using Cox' proportional hazard models and are presented as Hazard Ratios (HRs) with 95% confidence intervals (CIs). The 5-year mortality rates for women with ovarian cancer were calculated according to comorbidities diagnosed within one year after the cancer diagnosis and for comorbidities diagnosed between 1 and 5 years from cancer diagnosis. For the control group the mortality rates were calculated for comorbidities diagnosed within one year after the index date and for comorbidities diagnosed between 1 and 5 years from the index date. P-values for differences in HR between women with ovarian cancer and controls were calculated with a significance level of 0.05. For the women with ovarian cancer, we estimated the 5-year mortality rate by occurrence of comorbidity and stratified by histology group (high risk/medium risk epithelial tumors) and by age at ovarian cancer diagnosis (≤59 years/N60 years). In all adjusted analyses we included age at cancer diagnosis as a continuous variable in the models. The study was approved by one of the regional ethical boards at Karolinska Institutet, Stockholm.

Table 1 Numbers and survival rate of women with ovarian cancer by age at diagnosis, time period, histology and comorbidity. 1-Year survival N All Age at cancer diagnosis (years) –49 50–69 70–

(%)

11,139

5-Year survival

N

(%)

N

(%)

9193

(82.5)

5201

(46.7)

1867 5443 3829

(16.8) (48.9) (34.4)

1739 4782 2672

(93.1) (87.9) (69.8)

1270 2731 1200

(68.0) (50.2) (31.3)

Year of ovarian cancer diagnosisa 1993–1996 3438 1997–2001 3715

(30.9) (33.4)

2728 3120

(79.3) (84.0)

1437 2024

(41.8) (43.7)

Histologic type Epithelial, high riskb Epithelial, medium riskc Germ cell tumor Stroma cell tumor Sarcomas Other

8567 1504 146 452 202 268

(76.9) (13.5) (1.3) (4.1) (1.8) (2.4)

7078 1220 140 437 115 203

(82.6) (81.1) (95.9) (96.7) (56.9) (75.7)

3638 876 130 390 43 124

(42.5) (58.2) (89.0) (86.3) (21.3) (46.3)

Comorbidity Hematologic complications Thromboembolic events Cardiovascular disease Infectious disease Diabetes mellitus Hypertension

1460 929 1324 1277 565 1244

(13.1) (8.3) (11.9) (11.5) (5.1) (11.2)

– – – – – –

– – – – – –

381 209 424 390 190 525

(26.1) (22.5) (32.0) (30.5) (33.6) (42.2)

a Women with ovarian cancer diagnosis after 2001 were excluded due to incomplete survival data. b High risk epithelial tumors included serous and endometroid adenocarcinomas, anaplastic, small cell and poorly differentiated carcinomas. c Medium risk epithelial tumors included mucinous adenocarcinomas, clear cell carcinomas, and mesonefroid cancers.

300

K. Stålberg et al. / Gynecologic Oncology 133 (2014) 298–303

Results In total, 11,139 patients with invasive ovarian cancer were identified. Median age at diagnosis was 63 years and median survival was 653 days. Table 1 illustrates one and five year relative survival according to age, year of diagnosis, histological subtype and comorbidity. Older women had a shorter median survival than younger. The differences between the age groups were most apparent when assessing the proportions still alive 5 years after their diagnosis. Patients with ovarian sarcomas had the poorest 5-year survival rate (21%); followed by high risk epithelial and medium risk epithelial tumors. Patients with germ cell and stroma cell tumors had a high 5-year survival rate of 86–89%. The most frequent comorbidity was hematologic complications (13%) followed by cardiovascular disease and infections. Patients diagnosed with thromboembolism or hematologic complications had the poorest 5-year survival (22.5 and 26.1% respectively, Table 1). In Table 2, the relative 5-year mortality ratio by disease group is presented for women with and without ovarian cancer. For women with ovarian cancer, all comorbidity diagnoses increased mortality significantly except hypertension if diagnosed within one year after cancer diagnosis. Thromboembolism was the most hazardous (HR = 7.83 if diagnosed 1–5 years after cancer and HR = 1.95 within one year after cancer), followed by hematologic complications (HR = 7.11 and 1.84 respectively). When analyzing the separate diagnoses included in hematologic complications, thrombocytopenia was the most hazardous the first year after cancer diagnosis (HR = 5.50) and anemia after 1–5 years (HR = 7.54).

Also for women without OC, the mortality was higher when diagnosed with any of the selected comorbidities and the HRs were generally higher than among women with OC within 1 year of cancer diagnosis. Cardiovascular disease, diabetes and hypertension had a larger impact on 5-year mortality for women without OC than women with OC in all analyses. For women with less aggressive tumors (medium risk epithelial tumors) comorbidity had in general a larger impact on mortality than in women with highly aggressive tumors with statistically significant differences for thromboembolism, hematologic complications and infections (Table 3). The occurrence of hypertension only increased mortality significantly if diagnosed more than one year after cancer diagnosis (Table 3). In Table 4, mortality hazard ratios are presented by age at cancer diagnosis. Hematologic and thromboembolic complications increased the hazard ratios for women younger than 59 years of age more than 9-fold if diagnosed between 1 and 5 years after cancer diagnosis. For older women, the hazard ratios were also increased mainly for comorbidities diagnosed 1 to 5 years after the cancer diagnosis, though to a lesser extent than in the younger age group. For diabetes and hypertension there was no statistically significant difference between age groups. Discussion In this nationwide population-based study assessing impact of comorbidities in ovarian cancer patients, we found that all evaluated diseases increased mortality. The impact on mortality was most pronounced in women up to 59 years of age, women with less aggressive

Table 2 Relative 5-year mortality according to disease groups in 11,139 ovarian cancer (OC) patients and 55,687 age-matched women without ovarian cancer. Women with ovarian cancer

P-valuea

Women without ovarian cancer

Crude rate/1000 person yearsb

HRc

95% CI

Crude rate/1000 person yearsb

HRd

95% CI

Hematologic complications Exposure within one year after OCe, all Anemia Leucopenia Myelodyspastic syndrome Trombocytopenia Exposure one to five years after OCf, all Anemia Leucopenia Myelodyspastic syndrome Trombocytopenia

447.8 612.5 388.8 521.6 931.2 428.9 608.5 392.9 549.6 614.0

1.84 2.30 1.62 1.92 5.50 7.11 7.54 5.67 4.59 5.50

1.69–2.00 2.07–2.56 1.29–2.03 0.91–4.06 3.93–7.70 6.23–8.12 6.56–8.67 4.10–7.85 1.43–14.73 3.93–7.70

139.0

3.64

3.02–4.39

171.8

4.65

3.92–5.52

b0.001

Thromboembolic events Exposure within one year after OC Exposure one to five years after OC

679.7 529.6

1.95 7.83

1.75–2.18 6.78–9.05

132.3 146.6

2.53 4.95

2.00–3.20 4.04–6.05

b0.001

Cardiovascular disease Exposure within one year after OC Exposure one to five years after OC

597.1 319.8

1.44 1.76

1.31–1.58 1.46–2.11

104.1 120.0

2.89 3.46

2.65–3.15 3.13–3.83

b0.001

Infectious disease Exposure within one year after OC Exposure one to five years after OC

471.9 374.9

1.48 5.28

1.34–1.63 4.60–6.08

171.3 173.4

3.73 4.56

3.29–4.23 4.04–5.16

b0.001

Diabetes mellitus Exposure within one year after OC Exposure one to five years after OC

366.0 282.0

1.32 2.84

1.16–1.50 2.14–3.78

84.4 109.1

2.94 3.32

2.58–3.36 2.85–3.88

b0.001

Hypertension Exposure within one year after OC Exposure one to five years after OC

342.3 217.4

1.05 1.50

0.96–1.15 1.18–1.91

50.9 79.9

2.14 2.02

1.90–2.41 1.76–2.32

b0.001

a b c d e f

P-value for difference in HR between women with ovarian cancer and women without ovarian cancer for the whole study period. The calculated rate was crude and was not adjusted for attained age. Therefore the rate was not comparable with the hazard ratios. The hazard ratios (HR) were calculated using women with ovarian cancer and without the studied comorbidity as the reference. The hazard ratios (HR) were calculated using women without ovarian cancer and without the studied comorbidity as the reference. The studied comorbidity was diagnosed within one year after the ovarian cancer diagnosis and the corresponding interval was used for women without ovarian cancer. The studied comorbidity was diagnosed between one and five years after the ovarian cancer diagnosis and the corresponding interval was used for women without ovarian cancer.

K. Stålberg et al. / Gynecologic Oncology 133 (2014) 298–303

301

Table 3 Relative 5-year mortality according to comorbidity diagnosis in ovarian cancer (OC) patients by histology groups. High risk epithelial tumorsa

Medium risk epithelial tumorsb

P-valuec

Crude rate/1000 person years

HRd

95% CI

Crude rate/1000 person years

Hematologic complications Within 1 year after OC 1–5 year after OC

422.7 473.4

1.74 6.12

1.59–1.90 5.37–6.98

663.8 311.8

2.92 9.24

2.25–3.80 5.58–15.31

0.002

Thromboembolic events Within 1 year after OC 1–5 year after OC

671.1 577.7

2 6.78

1.79–2.25 5.90–7.80

815.5 453.5

3.64 16.67

2.74–4.84 10.45–26.57

b0.001

Cardiovascular disease Within 1 year after OC 1–5 year after OC

607.0 365.6

1.58 2.45

1.43–1.74 2.08–2.87

497.0 221.3

1.39 5.69

1.02–1.90 3.55–9.13

0.657

Infectious disease Within 1 year after OC 1–5 year after OC

448.8 437.6

1.55 5.24

1.40–1.72 4.58–5.98

604.3 236.7

1.89 8.81

1.41–2.53 5.73–13.57

0.047

Diabetes mellitus Within 1 year after OC 1–5 year after OC

357.0 340.2

1.29 3.06

1.13–1.48 2.34–4.00

575.9 113.4

1.65 2.37

1.11–2.45 0.79–7.08

0.780

Hypertension Within 1 year after OC 1–5 year after OC

336.3 243.4

1.05 1.35

0.95–1.16 1.06–1.74

422.9 127.7

1.14 3.76

0.85–1.54 1.95–7.23

0.499

a b c d

HRd

95% CI

High risk epithelial tumors included serous and endometroid adenocarcinomas, anaplastic, small cell and poorly differentiated carcinomas. Medium risk epithelial tumors included mucinous adenocarcinomas, clear cell carcinomas, and mesonefroid cancers. P-value for difference in HR between women with high risk tumors and women without medium risk tumors for the whole study period. The hazard ratios (HR) were calculated using women with ovarian cancer of the same histology group and without the studied comorbidity as the reference.

tumors and if the comorbidity was diagnosed more than one year after cancer diagnosis. Thromboembolism, hematologic complications and infections were the most hazardous diseases increasing the mortality 5 to10-fold if diagnosed more than a year after the cancer diagnosis. To our knowledge, we have the largest dataset published on comorbidity and ovarian cancer mortality based on information from population based registers. A very large study population is a prerequisite to achieve power enough to perform this kind of study with a quite rare

cancer diagnosis, separate comorbidity groups and subgroup analyses. With a smaller study population, the results often get to general or lack statistical significance. In some previous studies, comorbidities were grouped and indexed after their assumed impact on the outcome [25]. Such indices might be of value when comparing study populations, however, in clinical practice one might want to know how a specific disease or complication might influence the prognosis, making grouping of various diseases into an index less useful.

Table 4 Relative 5-year mortality according to comorbidity diagnosis in ovarian cancer (OC) patients by age at cancer diagnosis. ≤59 year at OC diagnosis

P-valuea

N60 year at OC diagnosis b

b

Crude rate/1000 person years

HR

95% CI

Crude rate/1000 person years

HR

95% CI

Hematologic complications Within 1 year after OC 1–5 year after OC

303.8 354.3

2.26 9.17

1.93–2.65 7.49–11.24

529.7 482.4

1.67 5.20

1.51–1.85 4.44–6.08

0.001

Thromboembolic events Within 1 year after OC 1–5 year after OC

530.4 454.1

2.59 9.04

2.09–3.20 7.27–11.25

743.3 575.1

2.01 6.48

1.78–2.26 5.50–7.64

b0.001

Cardiovascular disease Within 1 year after OC 1–5 year after OC

352.3 278

1.83 3.67

1.34–2.49 2.61–5.18

624.8 326.8

1.54 2.47

1.40–1.69 2.09–2.91

0.018

Infectious disease Within 1 year after OC 1–5 year after OC

281.4 339

2.01 7.33

1.68–2.42 6.00–8.97

572.4 398.8

1.48 4.61

1.33–1.65 3.93–5.41

b0.001

Diabetes mellitus Within 1 year after OC 1–5 year after OC

202.4 199.3

1.38 2.68

1.00–1.91 1.65–4.36

411.4 312.7

1.32 3.13

1.15–1.51 2.30–4.25

0.384

Hypertension Within 1 year after OC 1–5 year after OC

145.7 160.4

1.06 1.77

0.82–1.37 1.03–3.03

385.9 233.2

1.04 1.49

0.95–1.15 1.17–1.91

0.311

a b

P-value for difference in HR between women ≤59 year at OC diagnosis and women N60 years at diagnosis. The hazard ratios (HR) were calculated using women with ovarian cancer of the same age group and without the studied comorbidity as the reference.

302

K. Stålberg et al. / Gynecologic Oncology 133 (2014) 298–303

Another strength in this study is that data on diagnoses in The Swedish Patient Register is continuously registered by the treating doctors and the information is comprehensive and trustworthy and therefore all forms of recall bias are excluded. The routine of registering diagnosis is standardized nationwide in Sweden thus no regional differences occur. In addition, we have a long follow-up period in the CaCom database making comparison over time possible. A limitation when using pre-collected information from registers is often the lack of detailed information on covariates such as treatment and socio-economic factors. We did not have access to data on FIGO stage of the diagnosed cancer which is commonly used in epidemiological studies, yet it is known that histology is closely associated with cancer stage, and around 80% of serous ovarian cancers are diagnosed in stage 3–4 [26]. To approximate the influence of the malignant potential of the ovarian cancer, we performed subgroup analyses according to histological subtype in epithelial tumors. The tumor types in the medium risk group (mucinous adenocarcinomas, clear cell carcinomas, and mesonefroid cancers) are more often diagnosed in an early stage than the high risk tumors (seropapillar, endometroid and poorly differentiated tumors) [4]. In line with previous results we found a larger impact on mortality in women with less aggressive tumors [18]. Presumably, cancer related mortality is less significant in this group thereby allowing other conditions to have an effect on mortality ratios. The same pattern was seen when we compared younger and older women, as we found higher mortality HRs for women under the age of 59 year for most of the evaluated co-morbidities. This pattern has also been seen in other cancer diagnoses [3]. Accordingly it seems as comorbidity has a greater impact on mortality in women who had a more favorable situation at the time of the cancer diagnosis. Thromboembolism, hematologic diagnoses and infections are conditions with severe consequences for cancer patients and one should recall that these complications might be signs of recurrence of cancer and/or adverse effects of surgical or medical treatment. In this large study, it was not possible to scrutinize the medical records to obtain such information, however, when comparing with data for women without OC, the impact on mortality seems to be approximately the same for e.g. thromboembolism, speaking against a major bias in the cancer group. In some previous studies patients undergoing surgery actually show a decreased risk of thromboembolism maybe due to use of prophylaxis and/or patient selection bias [18,24]. Cardiac disease and ovarian cancer are sparsely studied in the literature, but some chemotherapeutic agents might affect cardiac function [19,20,21]. Because of the high mortality associated with cardiac infarction, we did not include that diagnosis among cardiovascular diseases. We found that cardiac disease increased 5-year mortality in women with OC with a HR of 1.76 if diagnosed after 1–5 year. It is known that the long-time cardiotoxic effect of chemotherapy can appear several years after the treatment [19]. Hypertension had small effect on mortality, and the results did not differ between younger and older women. The occurrence of diabetes mellitus had a moderate impact on survival and the results were rather unaffected by age or tumor histology. These findings have some support in the literature [1,9]. When comparing women with and without OC according to comorbidity one should acknowledge that cancer patients generally are under closer supervision and that other diseases will presumably be diagnosed at an earlier stage. Also less severe diseases in cancer patients might be recorded in connection with hospital admissions, which might not be the case for those not diagnosed with cancer. The Swedish Patient Register does not include diagnoses from primary care or from outpatient care for the time period, when the study was conducted. As risks for mortality by comorbidity in women with ovarian cancer were assessed in relation to women without cancer, an underestimation of less severe conditions, such as hypertension, in women without a cancer diagnosis would increase the risk estimates. This effect would not be apparent for more emergent diseases as illustrated by the effect on mortality by thromboembolism. Thromboembolism will most likely in both cancer

patients and those free of cancer be treated at the hospital. The results also show that thrombosis affect the mortality risk more in women with less aggressive tumors and in younger age. The findings in this study indicate that comorbidity in ovarian cancer patients, especially thrombosis, infections and hematologic complications has significant impact on survival. It is of great importance that these conditions are prevented or, if this is not possible, properly diagnosed and treated. After major surgery for ovarian cancer, patients should receive long time prophylaxis with low molecular weight heparin (LMWH), and prophylaxis should also be considered for women with large tumor burden without surgery [13]. In the subanalyses of the separate hematologic complications, thrombocytopenia was the most hazardous the first year and anemia 1–5 years after OC diagnosis and a great attention must be paid to prevent, diagnose and treat this common conditions. The use of Granocyte-colony stimulating factor (G-CSF) to prevent neutropenia in patients with solid tumors is important to reduce fatal infections and is shown to be cost-efficient [10,11]. Our results demonstrate that women with an otherwise more positive prognosis (younger and with less aggressive tumor) are at risk for increased mortality due to comorbidity and it is of uttermost importance that this group is properly informed of risks, and symptoms should not be neglected in follow-up programs. Conclusion The results from this large nationwide population-based register study show that common diagnoses in ovarian cancer patients, such as thromboembolism, hematologic complications and infections increased mortality significantly. The impact of comorbidity on mortality was most pronounced in subgroup analyses where the overall cancer mortality is lower i.e.; more than one year after cancer diagnosis, in women with less aggressive tumors and younger women. The occurrence of diabetes mellitus and hypertension had less impact on mortality among women with ovarian cancer. Conflict of interest statement The authors declare that there are no conflicts of interest.

Acknowledgments A special thanks to Fredrik Granath, PhD, Centre for Pharmacoepidemiology, Department of Medicine, Karolinska Institutet, for epidemiological and statistical advice and Bengt Tholander, MD, PhD, Department of Oncology, Radiology and Clinical Immunology, University Hospital, for support in managing SNOMED codes and oncology diagnoses. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ygyno.2014.02.024. References [1] Bakhru A, Buckanovich RJ, Griggs JJ. The impact of diabetes on survival in women with ovarian cancer. Gynecol Oncol 2011;121:106–11. [2] Barlow L, Westergren K, Holmberg L, Talback M. The completeness of the Swedish Cancer Register: a sample survey for year 1998. Acta Oncol 2009;48:27–33. [3] Berglund A, Garmo H, Tishelman C, Holmberg L, Stattin P, Lambe M. Comorbidity, treatment and mortality: a population based cohort study of prostate cancer in PCBaSe Sweden. J Urol 2011;185:833–9. [4] Bjorge T, Engeland A, Hansen S, Trope CG. Prognosis of patients with ovarian cancer and borderline tumours diagnosed in Norway between 1954 and 1993. Int J Cancer 1998;75:663–70. [5] Cloven NG, Manetta A, Berman ML, Kohler MF, DiSaia PJ. Management of ovarian cancer in patients older than 80 years of Age. Gynecol Oncol 1999;73:137–9. [6] Elit LM, Bondy SJ, Paszat LP, Holowaty EJ, Thomas GM, Stukel TA, et al. Surgical outcomes in women with ovarian cancer. Can J Surg 2008;51:346–54.

K. Stålberg et al. / Gynecologic Oncology 133 (2014) 298–303 [7] Engholm GFJ, Christensen N, Bray F, Gjerstorff Marianne L, Klint Å, Køtlum Jóanis E, et al. NORDCAN: Cancer incidence, mortality, prevalence and prediction in the Nordic countries. Danish Cancer Society; 2010. [8] Fader AN, von Gruenigen V, Gibbons H, Abushahin F, Starks D, Markman M, et al. Improved tolerance of primary chemotherapy with reduced-dose carboplatin and paclitaxel in elderly ovarian cancer patients. Gynecol Oncol 2008;109:33–8. [9] Ferriss JS, Ring K, King ER, Courtney-Brooks M, Duska LR, Taylor PT. Does significant medical comorbidity negate the benefit of up-front cytoreduction in advanced ovarian cancer? Int J Gynecol Cancer 2012;22:762–9. [10] Freyer G, Jovenin N, Yazbek G, Villanueva C, Hussain A, Berthune A, et al. Granocytecolony stimulating factor (G-CSF) has significant efficacy as secondary prophylaxis of chemotherapy-induced neutropenia in patients with solid tumors: results of a prospective study. Anticancer Res 2013;33:301–7. [11] Fust K, Li X, Maschio M, Barron R, Weinstein MC, Parthan A, et al. Cost-effectiveness of prophylaxis treatment strategies for febrile neutropenia in recurrent ovarian cancer patients. Value Health 2013;16:A140. [12] Klint A, Tryggvadottir L, Bray F, Gislum M, Hakulinen T, Storm HH, et al. Trends in the survival of patients diagnosed with cancer in female genital organs in the Nordic countries 1964–2003 followed up to the end of 2006. Acta Oncol 2010;49:632–43. [13] Lyman GH, Khorana AA, Kuderer NM, Lee AY, Arcelus JI, Balaban EP, et al. Venous thromboembolism prophylaxis and treatment in patients with cancer: American society of clinical oncology clinical practice guideline update. J Clin Oncol 2013;31:2189–204. [14] Maas HA, Kruitwagen RF, Lemmens VE, Goey SH, Janssen-Heijnen ML. The influence of age and co-morbidity on treatment and prognosis of ovarian cancer: a populationbased study. Gynecol Oncol 2005;97:104–9. [15] Markman M, Lewis Jr JL, Saigo P, Hakes T, Rubin S, Jones W, et al. Impact of age on survival of patients with ovarian cancer. Gynecol Oncol 1993;49:236–9.

303

[16] Olaitan A, Mocroft A, Jacobs I. Patterns in the incidence of age-related ovarian cancer in South East England 1967–1996. BJOG 2000;107:1094–6. [17] Ries LA. Ovarian cancer. Survival and treatment differences by age. Cancer 1993;71: 524–9. [18] Rodriguez AO, Wun T, Chew H, Zhou H, Harvey D, White RH. Venous thromboembolism in ovarian cancer. Gynecol Oncol 2007;105:784–90. [19] Sheppard RJ, Berger J, Sebag IA. Cardiotoxicity of cancer therapeutics: current issues in screening, prevention, and therapy. Front Pharmacol 2013;4:19. [20] Shrum KJ, Gill SE, Thompson LK, Blackhurst DW, Puls LE. New-onset congestive heart failure with gemcitabine in ovarian and other solid cancers. Am J Clin Oncol 2013 [in press]. http://dx.doi.org/10.1097/COC.0b013e31827b459a. [21] Smith LA, Cornelius VR, Plummer CJ, Levitt G, Verrill M, Canney P, et al. Cardiotoxicity of anthracycline agents for the treatment of cancer: systematic review and meta-analysis of randomised controlled trials. BMC Cancer 2010;10:337. [22] Stalberg K, Svensson T, Granath F, Kieler H, Tholander B, Lonn S. Evaluation of prevalent and incident ovarian cancer co-morbidity. Br J Cancer 2012;106:1860–5. [23] Storm HH, Engholm G, Hakulinen T, Tryggvadottir L, Klint A, Gislum M, et al. Survival of patients diagnosed with cancer in the Nordic countries up to 1999–2003 followed to the end of 2006. A critical overview of the results. Acta Oncol 2010;49:532–44. [24] Tateo S, Mereu L, Salamano S, Klersy C, Barone M, Spyropoulos AC, et al. Ovarian cancer and venous thromboembolic risk. Gynecol Oncol 2005;99:119–25. [25] Tetsche MS, Norgaard M, Jacobsen J, Wogelius P, Sorensen HT. Comorbidity and ovarian cancer survival in Denmark, 1995–2005: a population-based cohort study. Int J Gynecol Cancer 2008;18:421–7. [26] Tingulstad S, Skjeldestad FE, Halvorsen TB, Hagen B. Survival and prognostic factors in patients with ovarian cancer. Obstet Gynecol 2003;101:885–91.