Influence of specific comorbidities on survival after early-stage breast cancer

Abstract

Background: While comorbidity indices are useful for describing trends in survival, information on specific comorbidities is needed for the clinician advising the individual breast cancer patient on her treatment. Here we present an analysis of overall survival, breast cancer-specific mortality, and effect of medical adjuvant treatment among breast cancer patients suffering from 12 major comorbidities compared with breast cancer patients without comorbidities.

Material and methods: The study population was identified from the Danish Breast Cancer Cooperative Group and included 59,673 women without prior cancer diagnosed with early-stage breast cancer in Denmark from 1990 to 2008 with an estimated median potential follow-up of 14 years and 10 months. Information on comorbidity and causes of death was derived from population-based registries. Multivariable proportional hazards regression models were used to assess the effect of comorbidities on mortality, all-cause and breast cancer specific, using patients without comorbidity as reference.

Results: At breast cancer diagnosis, 16% of patients had comorbidities and 84% did not. Compared with the latter, the risk of dying from all causes was significantly increased for all types of comorbidity, but the risk of dying from breast cancer was significantly increased only for peripheral vascular disease, dementia, chronic pulmonary disease, liver, and renal diseases. Comorbidities diagnosed within 5 years of breast cancer diagnosis correlated with a greater risk of dying than comorbidities diagnosed more than 5 years before breast cancer diagnosis. With a few exceptions, the effect of adjuvant treatment on breast cancer mortality was similar among patients with and without comorbidity.

Conclusion: Breast cancer mortality was not significantly elevated for patients with prior myocardial infarction, congestive heart failure, cerebrovascular disease, connective tissue disease, ulcer disease, and diabetes. The similar effect of adjuvant treatment in patients with and without comorbidity underlines the importance of adhering to guideline therapy.

Introduction

Preexisting diseases at breast cancer diagnosis, comorbidity, may influence the prognosis after breast cancer in several ways. If the comorbidity involves organ failure, like compromised respiratory, cardiac, or renal function, curative treatment may not be possible leading to an increased risk of dying from breast cancer due to insufficient treatment. On the other hand, the risk of dying from the comorbidity may be so high that the patients may not live for sufficiently long time to benefit from breast cancer treatment even if they receive guideline therapy [1].

Most published reports on comorbidity have combined the diseases into one summary measure like the Charlson Comorbidity Index [2]. Several studies demonstrate that a high comorbidity score is associated with a reduced likelihood of receiving guideline therapy and an increased risk of dying from breast cancer as well as from all causes [1,3–5]. Yet, few studies have addressed associations between specific comorbid conditions and breast cancer prognosis. Such information may be helpful for the clinician advising the individual patient on her treatment. In a study of 64,034 patients with breast cancer diagnosed at a median age of 75 years identified from the Surveillance, Epidemiology, and End Results – Medicare database, Patnaik et al. [6] found that each of the 13 comorbid conditions examined was associated with decreased overall survival and increased mortality. Apart from prior cancer, diabetes was the condition with the highest prevalence, present in 13% of the patients. Compared with nondiabetic women, patients with breast cancer and preexisting diabetes are reported to have a greater risk of death, to present at later stages, and receive altered treatment regimens [7,8].

Based on data from the Danish Breast Cancer Cooperative Group (DBCG), we have previously reported that comorbidity as measured by Charlson’s Comorbidity Index at breast cancer diagnosis was an independent adverse prognostic factor for death after breast cancer [4]. The aim of this analysis was to examine the relationship between 12 major separate comorbidities derived from Charlson’s Comorbidity Index and overall survival, breast cancer specific mortality, and effect of medical adjuvant treatment compared with breast cancer patients without comorbidities.

Material and methods

We performed a population-based cohort study by linking the following Danish registers using the unique personal identification number: the Danish Breast Cancer Cooperative Group (DBCG), the Danish National Patient Register (NPR), the Central Population Register (CPR), and the Danish Register of Causes of Death (RCD).

From the database of the DBCG we identified 62,591 women diagnosed with early-stage breast cancer in Denmark from 1990 to 2008. The DBCG is a nationwide multidisciplinary group, which since 1977 has registered women diagnosed with primary invasive non-metastatic breast cancer with a completeness gradually improving to more than 95% [9]. The DBCG provided information on tumor characteristics and treatment.

Information on comorbidity was obtained from the NPR, which has registered in-patient diagnoses since 1977 and outpatient diagnoses since 1995 [10]. The International Classification of Disease (ICD)-codes were grouped into 12 categories, modified from the Charlson Comorbidity Index: myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic pulmonary disease, connective tissue disease, ulcer disease, liver disease (all grades), diabetes (all types), hemiplegia, and moderate to severe renal disease. We excluded 2910 patients with a prior cancer, leukemia or lymphoma, and eight patients with AIDS, leaving 59,673 patients available for analysis. Hospital contacts in a period from 10 years before and up to breast cancer diagnosis were included. ICD codes for identification of the comorbid diseases are listed in Supplementary Table I. Each patient can be registered with comorbidity in more than one category.

We grouped the diagnoses according to the time period in which the comorbidities were registered: within 5 years before the breast cancer diagnosis and from 10 to 5 years prior to the breast cancer diagnosis. Comorbidities can occur in both intervals for the same patient. The patients were followed from diagnosis to death, emigration or the end of December 2015 by linkage to the CPR. Information on cause of death was derived from RCD [11].

Statistical analysis

Endpoints were time to death, irrespective of cause (all-cause mortality and overall survival (OS)), and time to death from breast cancer (BC mortality), considering death from other causes as competing events. OS was calculated from time of breast cancer diagnosis using Kaplan–Meier estimates. BC mortality was calculated as cumulative incidence estimates. Follow-up time was quantified in terms of a Kaplan–Meier estimate of potential follow-up [12].

Univariate and multivariate regression analyses were performed, using Cox proportional hazards model for all-cause mortality and Fine-Gray proportional hazards subdistribution model for BC mortality. Comorbidities were included with one factor for each category, using patients with no comorbidity as reference. Factors included in the multivariable analyses were age (5-year categories), year of surgery, tumor size, lymph node status, lymphovascular invasion, invasion resection margin, histological type and grade, estrogen receptor (ER) status, menopausal status, and treatment (loco-regional and systemic). The assumption of proportional hazards was assessed by Schoenfeld residuals and by including time-dependent covariates in the model. Histology, ER status, and menopausal status did not fulfill the assumption and were included as stratification factors. Multivariate models were applied to investigate interaction between diabetes and vascular comorbidities in separate models, and to explore subgroups of comorbidity according to time and adjuvant treatment, respectively.

All p values are two-sided. Statistical analyses were done using SAS v9.4 (SAS Institute, Inc., Cary, NC, USA), and STATA IC 14 (StataCorp, College Station, TX, USA).

Results

The study population included 59,673 breast cancer patients with a median age of 61 years and with an estimated median potential follow-up of 14 years and 10 months. A total of 31,029 patients had died, 66% from breast cancer and 34% from other causes (Table 1). A third of the patients had a lumpectomy and about two thirds a mastectomy, while 5% had a biopsy only. For these, the histological diagnosis is often just stated as ‘carcinoma’ which explains why tumor size and histological type were unknown for about 6% of the patients. Seventy percent of the patients had ER positive tumors and 43% were node positive. Adjuvant medical treatment was not indicated for 27% of the patients because at the time of their diagnosis they were considered at low risk of recurrence, while 44% received chemotherapy or endocrine treatment alone or sequentially. For 29% the DBCG did not have information on given adjuvant therapy.

Table 1 Characteristics of 59,673 women diagnosed with early-stage breast cancer between 1990 and 2008 in Denmark.

	Characteristics of the study population
		Number of deaths
	Number of patients (%)	Breast cancer	Other causes
	59,673 (100)	20,381	10,648
Age at diagnosis
<40	2808 (5)	943	73
40–49	9875 (17)	2695	494
50–59	15,213 (25)	4396	1277
60–69	15,198 (25)	4855	2790
70–79	10,655 (18)	4456	3594
80+	5924 (10)	3036	2420
Tumor size
−10 mm	8257 (14)	1235	1454
11–20 mm	22,542 (38)	5638	4256
21–50 mm	22,267 (37)	9399	3874
51+ mm	2774 (5)	1797	346
Unknown	3833 (6)	2312	718
Histology and grade
Ductal I	14,192 (24)	3391	2948
Ductal II	19,649 (33)	6826	3320
Ductal III	10,587 (18)	4421	1276
Ductal unknown grade	1430 (2)	492	281
Lobular	6587 (11)	2320	1279
Other	4293 (7)	1011	953
Unknown	2935 (5)	1920	591
ER-status
Positive ≥10%	41,891 (70)	12581	7782
Poor <10%	11,734 (20)	4697	1547
Unknown	6048 (10)	3103	1319
Nodal status
Positive	25,943 (43)	11382	3355
Negative	31,079 (52)	7190	6798
Unknown	2651 (4)	1809	495
Surgery
Lumpectomy	19,047 (32)	3433	2389
Mastectomy	37,911 (64)	15042	7735
Biopsy only	2715 (6)	1906	524
Adjuvant therapy
None	15,915 (27)	3678	3094
Chemotherapy	7871 (13)	2723	465
Chemo- and endocrine treatment	4692 (8)	644	128
Endocrine alone	13,943 (23)	4302	2033
Unknown	17,302 (29)	9034	4928

Table 2 gives an overview of the 12 categories of comorbidity in relation to 5- and 10-year OS and BC mortality. The OS was 79% after 5 years and 64% after 10 years among 49,928 patients (84%) with no comorbidity registered before the breast cancer diagnosis. The presence of any of the comorbidities reduced OS, most pronounced for dementia with a 5-year OS of 28% and a 10-year OS of 9%. Similarly for BC mortality it was 18% at 5 years and 28% at 10 years among patients without comorbidity compared with 48% and 57% among women with dementia. The most common comorbidities were chronic pulmonary disease, present in 4.2%, followed by cerebrovascular disease (3.6%) and diabetes (3.1%).

Table 2. Comorbidity among 59,673 Danish women 0–5 and 5–10 years before diagnosis of breast cancer (BC) in relation to 5- and 10-year overall survival (OS) and breast cancer specific mortality (BCM).

	No. of patients	Disease before BC		No of deaths		5-year		10-year
		≤5 years	>5 years^a	BC	Other causes	OS % (95% CI)	BCM % (95% CI)	OS % (95% CI)	BCM % (95% CI)
Total	59673
None	49928			16616	7660	78.9 (78.5;79.2)	17.9 (17.6;18.2)	63.9 (63.5;64.3)	28.0 (27.6;28.4)
Myocardial infarction	831	562	322	351	305	56.7 (53.2;60.0)	28.6 (25.6;31.7)	34.3 (31.0;37.6)	39.2 (35.8;42.5)
Congestive heart failure	1168	958	302	502	511	42.4 (39.6;45.2)	33.5 (30.8;36.2)	18.3 (16.1;20.7)	42.5 (39.6;45.3)
Peripheral vascular disease	1076	837	365	430	397	54.6 (51.6;57.5)	28.9 (26.2;31.7)	30.1 (27.3;33.0)	38.2 (35.3;41.2)
Cerebrovascular disease	2154	1612	759	889	726	53.3 (51.2;55.4)	29.9 (27.9;31.8)	32.6 (30.5;34.6)	39.3 (37.2;41.4)
Dementia	369	325	64	211	132	28.1 (23.6;32.8)	48.0 (42.8;53.0)	09.3 (06.5;12.8)	56.8 (51.5;61.8)
Chronic pulmonary disease	2497	2018	888	901	764	62.0 (60.1;63.9)	24.4 (22.8;26.2)	40.4 (38.4;42.4)	34.2 (32.3;36.1)
Connective tissue disease	1236	928	563	428	329	66.7 (64.0;69.2)	22.5 (20.2;24.9)	46.1 (43.1;48.9)	33.0 (30.3;35.7)
Ulcer disease	1169	787	435	458	371	60.8 (58.0;63.6)	26.7 (24.2;29.3)	39.3 (36.4;42.2)	36.2 (33.4;39.0)
Liver disease	340	253	131	129	115	54.4 (49.0;59.5)	27.9 (23.3;32.8)	33.7 (28.5;39.0)	36.3 (31.1;41.5)
Diabetes	1843	1625	744	750	599	57.2 (54.9;59.4)	28.2 (26.2;30.3)	35.6 (33.3;37.8)	38.1 (35.8;40.3)
Hemiplegia	71	55	20	31	21	50.7 (38.6;61.6)	33.8 (23.0;44.9)	32.3 (21.5;43.7)	42.1 (30.0;53.7)
Renal disease	328	255	112	122	109	50.0 (44.5;55.3)	30.8 (25.9;35.8)	33.6 (28.4;38.8)	35.5 (30.4;40.7)

BC: breast cancer; OS: overall survival; BCM: breast cancer mortality; CI: confidence interval.

^a ≤10 years.

The results in Table 2 are as observed without any statistical modeling. Table 3 outlines the hazard ratio (HR) of dying from all causes and from breast cancer comparing patients with comorbidities with patients without comorbidities before breast cancer diagnosis. Adjustment for age at diagnosis reduced all HRs whereas adjustment for other factors (year of surgery, tumor size, lymph node status, lymphovascular invasion, resection margin, histological type and grade, ER status, menopausal status, treatment) resulted in minor changes. The HRs of dying from all causes remained statistically significantly increased for all 12 categories of comorbidity while the HR of dying from breast cancer also was significantly increased for peripheral vascular disease, dementia, chronic pulmonary disease, liver, and renal diseases, but not for myocardial infarction, congestive heart failure, cerebrovascular disease, connective tissue disease, ulcer disease, and diabetes.

Table 3. Multivariable analysis of the hazard ratio (HR) of death from all causes and from breast cancer (BC) for 12 comorbidities with breast cancer patients without comorbidities as reference category.

	All cause mortality HR (95% CI)			BC mortality HR (95% CI)
	Unadjusted	Age adjusted^a	Fully adjusted^b	Unadjusted	Age adjusted^a	Fully adjusted^b
Myocardial infarction	1.52 (1.40;1.65)	1.20 (1.11;1.30)	1.23 (1.14;1.33)	1.20 (1.07;1.35)	1.08 (0.96;1.21)	1.10 (0.98;1.24)
Congestive heart failure	2.22 (2.07;2.37)	1.42 (1.33;1.52)	1.42 (1.33;1.52)	1.28 (1.15;1.41)	0.99 (0.90;1.10)	0.99 (0.89;1.10)
Peripheral vascular disease	1.69 (1.57;1.81)	1.43 (1.33;1.53)	1.52 (1.42;1.64)	1.19 (1.07;1.32)	1.06 (0.96;1.18)	1.13 (1.01;1.27)
Cerebrovascular disease	1.81 (1.72;1.91)	1.30 (1.23;1.37)	1.30 (1.23;1.37)	1.26 (1.17;1.35)	1.05 (0.97;1.13)	1.05 (0.97;1.13)
Dementia	3.47 (3.12;3.87)	1.79 (1.61;2.00)	1.61 (1.44;1.80)	2.13 (1.82;2.50)	1.51 (1.28;1.78)	1.33 (1.13;1.58)
Chronic pulmonary disease	1.48 (1.40;1.55)	1.44 (1.37;1.52)	1.58 (1.50;1.66)	1.08 (1.00;1.16)	1.05 (0.97;1.12)	1.18 (1.10;1.28)
Connective tissue disease	1.28 (1.19;1.38)	1.14 (1.06;1.22)	1.22 (1.13;1.31)	1.02 (0.93;1.13)	0.95 (0.86;1.05)	1.01 (0.90;1.12)
Ulcer disease	1.49 (1.39;1.59)	1.21 (1.13;1.30)	1.28 (1.19;1.37)	1.15 (1.04;1.27)	1.03 (0.94;1.14)	1.09 (0.98;1.21)
Liver disease	1.76 (1.55;2.00)	2.01 (1.77;2.28)	2.25 (1.98;2.56)	1.15 (0.95;1.38)	1.23 (1.02;1.48)	1.41 (1.15;1.71)
Diabetes	1.58 (1.49;1.67)	1.39 (1.31;1.47)	1.33 (1.26;1.41)	1.22 (1.13;1.32)	1.13 (1.04;1.22)	1.05 (0.96;1.14)
Hemiplegia	1.59 (1.21;2.09)	1.52 (1.15;2.00)	1.42 (1.08;1.87)	1.33 (0.90;1.95)	1.27 (0.86;1.88)	1.27 (0.87;1.86)
Renal disease	1.42 (1.25;1.62)	1.46 (1.28;1.66)	1.69 (1.48;1.92)	1.07 (0.88;1.31)	1.07 (0.87;1.31)	1.26 (1.02;1.56)

HR: hazard ratio; BC: breast cancer; CI: confidence interval.

^aIn 5-year categories.

^b Adjusted for age, year of surgery, tumor size, lymph node status, lymphovascular invasion, invasion resection margin, histological type and grade, hormone receptor status, menopausal status, treatment.

Since most of the mortality associated with diabetes results from chronic vascular complications, we examined BC mortality in patients with diabetes only, vascular disease only, and diabetes with vascular complications (Table 4). BC mortality was not significantly increased in women with diabetes without complications but women with peripheral vascular disease without diabetes had a HR = 1.17 (95%CI 1.04–1.33). There was a borderline significant interaction (p = .05) between diabetes and congestive heart failure with HR = 0.83 (95% CI 0.64–1.08) in women with both conditions. A similar reduction of HR was seen for diabetes and peripheral vascular disease but the interaction did not reach statistical significance (p = .11).

Table 4. Interaction between diabetes and vascular comorbidities on the hazard ratio of death from breast cancer with breast cancer patients without comorbidity as reference.

	Vascular morbidity No diabetes	Diabetes No vascular morbidity	Diabetes and Vascular morbidity
	HR (95% CI)	HR (95% CI)	HR (95% CI)	p test of interaction
Myocardial infarction	1.11 (0.97;1.26)	1.05 (0.96;1.15)	1.12 (0.82;1.51)	.82
Congestive heart failure	1.03 (0.92;1.16)	1.08 (0.99;1.18)	0.83 (0.64;1.08)	.05
Peripheral vascular disease	1.17 (1.04;1.33)	1.07 (0.98;1.17)	0.94 (0.68;1.30)	.11
Cerebrovascular disease	1.05 (0.96;1.14)	1.05 (0.96;1.14)	1.10 (0.88;1.38)	.97
Renal disease	1.21 (0.96;1.51)	1.04 (0.96;1.13)	1.59 (0.94;2.70)	.42

HR: hazard ratio; CI: confidence interval.

We examined whether there was any difference in all-cause mortality between more recent comorbidity occurring within 5 years of the breast cancer diagnosis and past comorbidity occurring 5–10 years before the breast cancer diagnosis (Figure 1). Although all HR-estimates were significantly increased, there was a trend of higher HRs for recent comorbidity, except for hemiplegia with HR = 1.23 (95% CI 0.90–1.69) if registered within 5 years and HR = 2.88 (1.75–4.72) for years 5–10 (p for heterogeneity .006). Different estimates were also seen for cerebrovascular disease with HR = 1.30 (95% CI 1.22–1.38) for years 0–5 and HR = 1.11 (95% CI 1.01–1.21) for years 5–10 (p for heterogeneity .008), chronic pulmonary disease with HR= 1.55 (95% CI 1.46–1.65) for years 0–5, and HR= 1.17 (95% CI 1.07–1.28) for years 5–10 (p for heterogeneity <.0001) and for ulcer disease with HR = 1.33 (95% CI 1.23–1.45) for years 0–5 and HR = 1.13 (95% CI 1.00–1.27) for years 5–10 (p for heterogeneity .03). For diabetes, the p value for heterogeneity was .046. We also examined BC mortality but the results were largely similar to all-cause mortality (data not shown).

Figure 1. Hazard ratio of death from all causes in relation to time period of comorbidity with breast cancer patients without comorbidity as reference.

With a few exceptions, the effect of adjuvant treatment on BC mortality was similar among patients with and without comorbidity (Figure 2). Patients suffering from dementia receiving chemotherapy had a four-fold higher risk of dying (HR = 4.05, 95% CI 1.72–9.52), but this estimate was based on only eight patients. Among patients receiving endocrine treatment only, those with chronic pulmonary disease had HR = 1.16 (95% CI 1.01–1.35), liver disease HR = 1.40 (95% CI 1.00–1.94), and diabetes HR = 1.21 (95% CI 1.03–1.42).

Figure 2. Hazard ratio of death from breast cancer in relation to receipt of adjuvant chemotherapy (CT) and adjuvant endocrine treatment (ET) with breast cancer patients without comorbidity as reference.

Discussion

The prevalence of comorbidity increases markedly with age from less than 10% in breast cancer patients aged less than 50 years to 40% for those aged 80 years or more [4]. The relatively low prevalence of comorbidity (16%) in our study reflects that the median age was 61 years. In the US Medicare Population which included only individuals 65 years or older, the prevalence of comorbidity was similar among cancer-free Medicare beneficiaries (32%), breast cancer patients (32%), and prostate cancer patients (31%), but higher among patients with colorectal cancer (41%) and lung cancer (53%) [13].

The effect of age was also evident when adjustments for age were made in the multivariable analysis, e.g. for congestive heart failure, an unadjusted HR = 2.22 (95% CI 2.07–2.37) for death from all causes was reduced to HR = 1.42 (95%CI 1.33–1.52) after adjustment for age (Table 3). Despite adjustment for all available potential confounding factors, the all-cause mortality remained significantly increased for all the 12 comorbidities examined. This finding is in agreement with those of other studies [3,6,7]. However, it was reassuring that BC mortality was not significantly elevated for patients with prior myocardial infarction, congestive heart failure, cerebrovascular disease, connective tissue disease, ulcer disease, and diabetes. It was equally reassuring that with a few exceptions, the effect of adjuvant treatment was similar for patients with and without comorbidity. This finding is new and underlines the importance of adhering to guideline therapy even among patients who suffer from comorbidity.

Most register-based studies from Denmark have examined comorbidity over a period of 10 years preceding the breast cancer diagnosis [4,14], while others have collected the information at diagnosis without specifying when the comorbidity occurred [5]. Our results suggest that more recent comorbidity, i.e., within 5 years, carries a greater risk of dying than comorbidity occurring more than 5 years before breast cancer except for hemiplegia where the reverse was seen. For diabetes, these findings confirm the observations by Goodwin et al. [15] of a pronounced effect on mortality of insulin-related variables during the first 5 years after diagnosis, but not thereafter. Such information is relevant for the clinician advising the individual patient on her treatment. Radiotherapy increases the risk of a myocardial infarction, particularly for left-sided breast cancers, and this risk is 3–5-fold increased for women with prior ischemic heart disease [16,17]. Chemotherapy with anthracyclines is also cardiotoxic [18] and the risk of heart disease may increase further by radiotherapy [19].

Information on diabetes relied in this material only on diagnoses reported to the NPR resulting in a prevalence of just over 3%. When other sources of information like measurements of blood glucose and diabetes medication were taken into account, the Danish Diabetes Registry reported prevalences of 3–6% for women aged 60 years between 1996 and 2010 with the prevalence peaking at 17% for women aged 80 years [20]. Diabetes is associated with microvascular complications such as kidney disease, which is a risk factor for macrovascular disease, i.e., atherosclerosis, myocardial infarction, heart failure, and cerebrovascular events [21]. We had expected that the presence of diabetes with such vascular complications would be associated with a higher BC mortality than diabetes without complications, but the data did not support this. On the contrary, HRs less than one were observed among diabetics with congestive heart failure and peripheral vascular disease although the estimates failed to reach statistical significance.

The strength of this study is that it was based on a prospective data collection in an entire population with valid information on tumor characteristics, given breast cancer-directed treatment, and all-cause mortality. Follow-up was 100% complete and the study population was followed for over 14 years. However, there are limitations. We had no information on potentially confounding factors, such as smoking or obesity. Data on given treatment were missing for 29% of the study population, partly explained by patients aged 75 years or older at diagnosis, who up to 2002 were not included in the national treatment guidelines [4]. Since the DBCG does not follow patients after a recurrence or a maximum of 10 years for recurrence-free patients [9], our only way to ascertain if deaths were due to breast cancer was to rely on information from death certificates. However, these are likely to be correct for patients dying with metastatic breast cancer.

Conclusion

This study confirmed that all-cause mortality was significantly increased for all the 12 comorbidities examined, but it also gave evidence that BC mortality was not significantly elevated for patients with prior myocardial infarction, congestive heart failure, cerebrovascular disease, connective tissue disease, ulcer disease, and diabetes. Our results suggest that more recent comorbidity, i.e., within 5 years, carries a greater risk of dying than comorbidity occurring more than 5 years before breast cancer. Finally, it was reassuring that with a few exceptions, the effectiveness of adjuvant treatment was similar for patients with and without comorbidity.

Disclosure statement

No potential conflict of interest was reported by the authors.