Support Care Cancer DOI 10.1007/s00520-015-2901-8

ORIGINAL ARTICLE

Port catheter versus peripherally inserted central catheter for postoperative chemotherapy in early breast cancer: a retrospective analysis of 448 patients L. Lefebvre 1 & E. Noyon 2 & D. Georgescu 3 & V. Proust 2 & C. Alexandru 4 & M. Leheurteur 4 & J. C. Thery 4 & L. Savary 4 & O. Rigal 4 & F. Di Fiore 4 & C. Veyret 4 & F. Clatot 4

Received: 15 May 2015 / Accepted: 9 August 2015 # Springer-Verlag Berlin Heidelberg 2015

Abstract Purpose We aimed to compare the complication rate between port catheters (PC) and peripherally inserted central catheters (PICC) for the administration of postoperative chemotherapy for breast cancer. Methods All patients treated from January 2010 to August 2012 at the Centre Henri Becquerel for early breast cancer requiring postoperative chemotherapy were retrospectively screened. The primary endpoint was the occurrence of a major complication related to the central venous catheter. Major complications were defined as any grade 3 event according to CTCAE 4.0, delay in chemotherapy >7 days, change of the device, life-threatening event, event requiring a hospitalization, or a prolongation of hospitalization. Results A total of 448 patients were included; 290 had a PC and 158 a PICC. Overall, 31 major complications related to the central venous catheter were observed: 13 for patients with a PC (4.5 %) and 18 for patients with a PICC (11.4 %). In univariate analysis, having a PICC was the only factor significantly associated with a higher risk of major complications (HR=2.83, p=0.0027). We observed a trend for a higher risk of major complications for patients older than 60 years or with

* F. Clatot [email protected] 1

Department of Radiation Oncology, Centre Henri Becquerel, 1 rue d’Amiens, 76038 Rouen, France

2

Department of Anesthesiology, Centre Henri Becquerel, 1 rue d’Amiens, 76038 Rouen, France

3

Department of Surgery, Centre Henri Becquerel, 1 rue d’Amiens, 76038 Rouen, France

4

Department of Medical Oncology, Centre Henri Becquerel, 1 rue d’Amiens, 76038 Rouen, France

BMI >25 (p=0.06). In multivariate analysis, having a PICC was the only predictive factor of major complications (HR= 2.89, p=0.004). Conclusions In univariate and multivariate analysis, having a PICC instead of a PC was the only predictive factor of devicerelated major complication. If confirmed prospectively by the NCT02095743 ongoing trial, this result might modify the management of adjuvant chemotherapy administration. Keywords Central venous catheter . Breast neoplasm . Chemotherapy . Adjuvant . Port catheter . Peripherally inserted central venous catheter (PICC)

Introduction Breast cancer is the most common cancer among women with 464,000 patients diagnosed in Europe in 2012 [1]. Usually, breast cancer is diagnosed at an early stage and treated by surgery and radiotherapy. Adjuvant chemotherapy is frequently proposed in cases of poor prognosis factors (like nodal invasion, high histological grade, tumor size, or triple negative cancers). This chemotherapy is based on 6 cycles (4 months) and significantly decreases the risk of recurrence [2, 3]. A central venous device is required for the administration of this adjuvant chemotherapy to avoid peripheral venous punctures and venous toxicities. To date, two devices can be used in this indication. Port catheter (PC) is the classical route of administering chemotherapy because it provides deep venous access and allows for iterative perfusions, even for years. Nevertheless, the insertion and removal of a PC are invasive and questionable when the use of the device is expected for a limited period. Initially, the peripherally inserted central catheter (PICC) was developed to provide deep venous access for parenteral nutrition and is commonly used in intensive care

Support Care Cancer

units or for pediatric patients [4, 5]. A PICC can be easily inserted and removed but needs maintenance at least once a week (StatLock® PICC Plus Usage, Bard Access Systems Inc., Salt Lake City, UT). Currently, PICCs are also used for chemotherapy [6, 7]. The same pattern of complications (mainly thromboembolic events and infections) is observed for both PCs and PICCs, and little evidence-based medicine is available for the choice between the two devices. Some retrospective studies have reported a higher rate of complications (in particular, thrombotic complications) among oncologic patients with a PICC [8–12]. Nevertheless, many methodological limitations might have misestimated this rate of complication: available studies included patients regardless of primary tumor site, stage of the disease, utilization of the device (supportive care, chemotherapy, antibiotherapy), and duration of use. Only two prospective trials have compared PC and PICC. The first trial was not randomized and included 116 patients with solid or hematologic malignancies [13]. The authors found 50 % of complications with PICCs and 34.7 % of complications with PCs, but complication types were not detailed in relation to the type of device. The second trial was randomized but limited to 70 patients [14]. The authors reported significantly more major complications with PICCs than with PCs (20 versus 6 %, respectively). However, the conclusion of this study is hardly applicable to the use of adjuvant chemotherapy because this study included 60 % of patients with metastatic diseases and various primitive sites of cancer (especially colorectal cancers and upper gastrointestinal and breast cancers), which might have impacted the rate of thromboembolic events. Finally, the latest Clinical Practical Guidelines of the American Society of Clinical Oncology could not recommend one device over the other for the administration of adjuvant chemotherapy for early breast cancer [15]. In our Cancer Care Centre, the PC and PICC have been used routinely for years for early breast cancer patients receiving adjuvant chemotherapy. The aim of this study was to retrospectively assess the rates of complications for these devices.

Patients and methods This retrospective study was conducted at the Centre Henri Becquerel, Rouen, France. All patients treated between January 2010 and August 2012 were screened. The inclusion criteria were female patients older than 18 treated for a Her2negative early breast cancer confirmed by a biopsy and requiring postoperative chemotherapy (as determined by our internal Breast Cancer Board) with 6 cycles of FEC 100 (5 fluorouracil 500 mg/m2, epirubicin 100 mg/m2, cyclophosphamide 500 mg/m2) or 3 cycles of FEC 100 followed by 3 cycles of docetaxel (D) (100 mg/m2). We excluded patients with a

history of thoracic radiation therapy, bilateral axillary node dissection, curative anticoagulant therapy at cancer diagnosis, INR>1.5, platelet count <60 G/L, TCA>1.5, renal failure (MDRD<60 mL/min/1.73 m2), chemotherapy contraindication, pregnancy or breast-feeding, and psychiatric disease. This study was approved by our Internal Review Board. Primary endpoint The primary endpoint was the occurrence of a major complication related to the central venous catheter. Complications were considered related to the central venous catheter if they met at least one of the following criteria: placement complications (pneumothorax on pulmonary X-ray, hematoma observed on ultrasound, arterial puncture, nerve damage, placement failure, local infection, or bacteremia before the first use of the device), local infection or catheter-related bacteremia (as defined by the French National Committee of nosocomial infections and care related infections, May 2007), dressing allergy, thrombosis on the catheter confirmed by one radiologic modality, pain related to the device assessed by visual analog scale and by need for analgesic treatment, infusion extravasation, accidental removal, and mechanical complications (catheter dislodgement, migration of the tip, reversal of the port, occlusion despite the use of heparin, or urokinase protocols). Complications were considered major if they met at least one of the following criteria: any grade 3 event according to Common Terminology Criteria for Adverse Event (CTCAE) 4.0, delay in chemotherapy >7 days, change of the device, life-threatening event, event requiring a hospitalization or a prolongation of hospitalization, and pain requiring morphine treatment. We recorded the complications observed between the device insertion and its removal. For PICCs, removal was performed just after the end of the 6th cycle of chemotherapy. Because PCs are not routinely removed at the end of chemotherapy, we censored the analysis at 6 months following the last cycle of chemotherapy. If a patient presented several major complications, only the first one was used for the statistical analysis of the primary endpoint. Event recordings were based on the medical consultation report, which was mandatory before each chemotherapy infusion, and on the hospital report in the event of a complication. Secondary endpoint The secondary endpoint was the occurrence of minor complications related to the device defined by complications related to the device according the previous criteria but not classified as major complication criteria. Complications were considered related to the central venous catheter if they met at least one of the following criteria: placement complications (pneumothorax on pulmonary X-ray, hematoma observed on

Support Care Cancer

ultrasound, arterial puncture, nerve damage, placement failure, local infection before the first use of the device), local infection, dressing allergy, pain related to the device assessed by visual analog scale and by need for analgesic treatment, infusion extravasation, accidental removal, and mechanical complications (catheter dislodgement, migration of the tip, reversal of the port, occlusion despite the use of heparin, or urokinase protocols). Complications were considered minor if they did not meet one of the following criteria: grade 3 event according to CTCAE 4.0, delay in chemotherapy >7 days, change of the device, life-threatening event, event requiring a hospitalization or a prolongation of hospitalization, and pain requiring morphine treatment. The data recording time was the same as for the major complications. Central venous access devices Two types of devices were used: PC or PICC. The choice of the device was at the discretion of the physician. PCs had two components: an injection port with a self-sealing silicone septum and a radiopaque 6 Fr catheter, from various manufacturers. PICCs were 4 Fr single-lumen polyurethane catheters from Bard Access Systems, Inc. Salt Lake City, Utah, USA. PCs were inserted in the operating room after local anesthesia. After ultrasound-guided puncture of the internal jugular vein or subclavian vein, the subcutaneous port was placed under the clavicle, and the device was tested. The procedure time was approximately 30 to 45 min. PICCs were inserted in the operating room. The basilica vein was punctured above the elbow under ultrasound guidance and after local anesthesia. The procedure time was approximately 15 to 30 min. Both procedures were performed under aseptic conditions, and radiographic controls were used during installation. The same day, the catheter position was checked with a chest X-ray. The position was considered valid if the tip of the catheter was between the lower part of the inferior vena cava and the right atrium. No patient was given prophylactic antibiotics or anticoagulation treatment. No maintenance was required for PCs. For PICCs, a weekly maintenance with dressing change and saline wash was performed by a home-care nurse. Statistical analysis Comparisons between the two groups of patients were performed using the chi-square test. The Kaplan-Meier method was used to determine the probability of occurrence of complications. The log rank test was used to test the differences between the curves. Multivariate analysis of predictive factors was performed according to Cox proportional-hazard regression with the backward method. Statistical analyses were performed using MedCalc® medical software (Ostend, Belgium).

Results Four hundred forty-eight women were retrospectively enrolled. The mean age at diagnosis was 54 years. The two groups were comparable except for the type of breast and node surgery. Patient characteristics are described in Table 1. Among the patients, 290 had a PC (64.7 %), and 158 (35.3 %) had a PICC. Thirty-one patients had at least one major complication induced by the central venous catheter, including 13 of the 290 patients with a PC (4.5 %) and 18 of the 158 patients with a PICC (11.4 %) (p=0.006, chi-square test). Two patients in the PICC arm had 2 major complications; one patient had 2 catheter thromboses related to 2 successive PICCs, and the other patient had a local infection of a PICC followed by a catheter thrombosis related to a second PICC, after replacement of the first PICC. Our analysis was restricted to the first occurrence of a major complication. The different types of major complications are reported in Table 2. These complications led to anti-coagulation treatment initiation in 18 cases (7 in the PC arm and 11 in the PICC arm), chemotherapy delay in 8 cases (4 in the PC arm and 4 in the PICC arm), hospitalization in 6 cases (2 in the PC arm and 4 in the PICC arm), and device replacements in 16 cases (6 in the PC arm and 10 in the PICC arm). Major complications occurred after a median number of 1 chemotherapy cycle (30 days after the catheter placement). A significantly higher rate of major complications was observed during the first 3 cycles of chemotherapy than during the last 3 cycles: 25 (81 %) major complications occurred before the fourth cycle vs 6 (19 %) after the fourth cycle (p=0.0005, chi-square test). In the univariate analysis, having a PICC was the only predictive factor significantly associated with a higher risk of device-related major complications, with a hazard ratio (HR) of 2.83 (95 % CI [1.3–6], p=0.0027). There was a trend for a higher risk of major complications for patients older than 60 years or with a BMI>25 (p=0.058 in both cases). The type of breast and node surgery was not a risk factor in the univariate analysis (Table 3). Figure 1 shows the probability of occurrence of major complications for patients having a PICC or a PC. In the multivariate analysis, having a PICC instead of a PC was the only predictive factor of device-related major complications (HR = 2.89, 95 % CI [1.4–8], p = 0.004) (Table 4). Overall, 37 patients had at least one minor complication induced by the central venous catheter, including 12 minor complications in patients with a PC (4.1 %) and 25 in patients with a PICC (15.8 %). The types of complications observed are described in Table 2. Minor complications occurred after a median number of 48 days and 2 cycles of chemotherapy. In the univariate analysis, PICC was the only predictive factor significantly associated with a higher risk of devicerelated minor complications, with HR =3.9 (95 % CI [2– 7.8], p<0.0001). Age older than 60 years, BMI≥25, and

Support Care Cancer Table 1 Primary characteristics of the patients Age (years) BMI (kg/m2)

Nodal invasion T stage

PC n=290

PICC n=158

Chi2 or Fischer test

<60 ≥60 <25 25 to 30

196 (68 %) 94 (32 %) 139 (48 %) 88 (30 %)

91 (58 67 (42 79 (50 48 (30

p=0.04

≥30 Yes No

63 (22 %) 169 (58 %) 121 (42 %)

31 (20 %) 80 (51 %) 78 (49 %)

T0 T1

1 (0.3 %) 170 (59 %)

0 102 (65 %)

T2

102 (35 %)

49 (31 %)

T3 T4

17 (5.8 %) 0 246 (85 %) 25 (9 %) 8 (3 %)

7 (4 %) 0 135 (85 %) 16 (10 %) 5 (3 %)

Histology

IDCa ILCb IDCa +ILCb Others

HRc expression

Presence Absence Yes No Lumpectomy Mastectomy Axillary node dissection Sentinel node procedure 6FEC 3FEC-3D

Her 2 Overexprimed Surgery Node surgery Chemotherapy a

%) %) %) %)

9 (3 %)

2 (1 %)

225 (78 %) 65 (22 %) 0 (0 %) 290 (100 %) 184 (63 %) 106 (37 %) 91 (31 %) 199 (69 %) 100 (34 %) 190 (66 %)

127 (80 %) 31 (20 %) 0 (0 %) 158 (100 %) 116 (73 %) 42 (27 %) 99 (63 %) 59 (37 %) 55 (35 %) 103 (65 %)

p=0.9

p=0.1 p=0.6

p=0.7

p=0.5 p=1 p=0.03 p<0.001 p=0.9

Invasive ductal carcinoma

b

Invasive lobular carcinoma

c

Hormonal receptors

chemotherapy regimen were not significantly associated with a higher risk of minor complications. In the multivariate analysis, having a PICC was the only predictive factor significantly associated with a higher risk of device-related minor complications, with HR=3.9 (95 % CI [2–7.8], p=0.0001).

Discussion Our study is the first to assess PICC and PC for adjuvant chemotherapy administration in breast cancer patients. In this cohort, PICC was the only risk factor of catheter-related major and minor complications. In the PC arm, 8.6 % of patients had at least one complication versus 27.2 % in the PICC arm. Among these complications, the major ones were also significantly more frequent in the PICC arm (4.5 vs 11.4 %). Thrombosis was the most frequent major complication, and overall, these complications occurred at the beginning of treatment. As treatment began with FEC, the majority of complications occurred during FEC administration. Only 6 patients

(19 %) had major complications during D cycles (fourth cycle and after). One might hypothesize that FEC cycles are associated with the occurrence of higher rates of complications than D cycles. However, among the 155 patients (34.6 %) who received 6 cycles of FEC instead of 3 cycles of FEC followed by 3 cycles of D, no increased complications were observed in univariate or multivariate analyses (Tables 3 and 4). Thus, this result suggests that the probability of complications decreases with time. It also suggests that PICC could be a risk factor of complications even for a short duration of use. Our study involves some bias due to its retrospective design. The most important is the missing data. In our institution, all patient records are computerized, which to some extent limits this bias. Concerning major complications, only very few data might be missing because most of these complications were managed in our institution, and complications managed by another facility (like an emergency room) were reported in detail during the systematic medical consultation performed before every chemotherapy administration. In contrast, it is likely that minor complications were not reported as systematically as major complications. Thus, the rate of minor

Support Care Cancer Table 2

Description of complications related to the device Port n=290

PICC n=158

Placement complications

1

5

Major

0

0

Minor Local infection/inflammation

1 4

5 10

Major Minor

1 3

2 8

Bacteremia (major)

1

1

Thrombosis (major) Septic thrombophlebitis (major)

6 1

10 1

Pain Major

2 1

1 0

Minor

1

1

Infusion extravasation Major

1 0

0

Minor Mechanical complications

1 8

7

Major

3

3

Minor Dressing allergy Major Minor

5 0 0 0

4 6 0 6

Others (lymphatic leakage) Major Minor Total Major

1 0 1 25 (8.6%) 13 (4.5%)

2 1 1 43 (27.2%) 18 (11.4%)

Minor

12 (4.1%)

25 (15.8%)

Fig. 1 Probability of occurrence of major complications related to the device

All the thrombosis observed needed an anticoagulation therapy and are therefore at least a grade 3 according to the CTCAE 4.0

complications might be underestimated and/or biased in our cohort. The complication rate of PC withdrawal might also be underestimated because only a few patients had their PC

removed at the time of data collection. One other important bias is the selection of the patients receiving one device over the other. The choice of the device was discussed between the physician and the patient. In our cohort, ages, type of breast, and node surgery were significantly different between the 2 groups: patients were older in PICC arm and two times as many patients with a PICC underwent axillary node dissection rather than a sentinel node procedure. Even if the age was only slightly different with a mean age at 53 years (DS=10) in the PC arm and at 55 years (DS=9.6) in the PICC arm, we cannot rule out that this small difference could play a role in the occurrence of complications. Concerning the type of breast or node surgery, these criteria were not taken into account in the choice of central access devices by the physician, and we can only explain this difference by hazard. We may hypothesize that these settings had a small impact on our results because the PICC were systematically inserted in the other side of axillary node dissection and patients with bilateral axillary Table 4 Pronostic factors of major complications related to the devices, multivariate analysis

Table 3 Pronostic factors of major complications related to the device, univariate analysis PC PICC

1 2.83 [1.3-6]

0.003

Age<60 years Age≥60 years BMI<25 (kg/m2) BMI≥25 6FEC 3FEC-3D Lumpectomy Mastectomy Sentinel node procedure Axillary lymph node dissection

1 1.9 [0.9–4] 1 2 [1–4.2] 1 1.7 [0.6–2.8] 1 1.6 [0.8–3.4] 1 1.4 [0.6–2.9]

0.058 0.058 0.45 0.2 0.4

HR 95 %CI

p

PC PICC

1 2.89 [1.4–8]

0.004

Age<60 years Age≥60 years BMI<25 (kg/m2) BMI≥25 6FEC 3FEC-3D Lumpectomy Mastectomy Sentinel node procedure Axillary lymph node dissection

1 0.5 [0.2–1.1] NIa

0.08

NI

NI

NI

NI

NI

NI

a

Not included by the backward model

NI

Support Care Cancer

dissection were excluded. Moreover, some publications based on large cohorts of patients found no difference for thrombotic or septic complications (other than wound infection) in women who underwent axillary lymph node dissection or sentinel lymph node procedure [16, 17]. Furthermore, in our study, the age and the type of breast or node surgery were not related to the risk of device-related complications, either in a univariate or multivariate analysis (Tables 3 and 4). In our institution, we started using PICCs in mid-2009, and data collection started in January 2010. Considering a potential bias due to the learning time for the insertion of a PICC, we did not find more complications in the first month of the study compared to the last months (data not shown). Our results are consistent with those reported in the literature. In the most recent prospective study, despite differences in the inclusion criteria between our study and their cohort of 70 patients, Patel et al. found a higher rate of major complications in the PICC arm (20 %) than in the PC arm (6 %). We found a comparable ratio between our two arms (4.5 % for PC and 11.4 % for PICC). The higher values of major complications observed in Patel et al. study might be due to high rates of gastrointestinal tumors included in their study which are usually associated to more thromboembolic complications than breast cancer [18–20]. Of note, Patel et al. reported that having a metastatic disease was not a risk factor of complications in multivariate analyses [14]. Patel et al. also reported that there was no difference in cost between PICCs and PCs because even if the PC insertion and removal were more expensive, this cost was counter-balanced by the cost of maintenance and complications for PICCs. Quality of life was also not different between the two groups. In another published prospective study, Kim et al. reported a higher rate of tip migration with PICCs versus PCs, but no difference in rate of infections, thrombosis or bleeding, even if the detailed rates are not available in their article. These findings might be explained by a lack of power of the study, as only 24 patients had a PICC (versus 92 patients who had a PC) and the PICC life span was very short (21 days versus 269 for PCs) [13]. One retrospective study published by Xing et al. in 2012 focused on thrombosis complications in 187 breast cancer patients with PICCs for the administration of chemotherapy (all stages of disease) and found a rate of thrombosis of only 2.1 %, concluding that PICCs are safe to use in this indication. In our study, we found 7 % of thrombotic complications in the PICC arm. Of note, Xing et al. included patients receiving chemotherapy by PICC even for a single day, and it was not specified if a thromboprophylaxis treatment was administered [21]. No antibiotics or anticoagulation prophylactic treatments were given in our cohort, and catheters were not impregnated with antibiotics or heparin. It has been shown that impregnated catheters were associated with a lower rate of catheter-

related infections [22, 23], but the use of impregnated catheters is controversial because of their cost, and they are recommended only for high risk patients, such as patients with bone marrow transplantation or leukemia [15]. Concerning the prophylactic use of systemic antibiotics, a recent review of the ASCO did not find a diminution in the risk of catheter-related bacteremia with prophylactic systemic antibiotics [15]. In contrast, in a Cochrane review, flushing or locking the central venous catheter with an antibiotic and heparin solution appeared to reduce Gram-positive catheter-related bacteremia in patients at risk of neutropenia, such as patients receiving chemotherapy, but the data were insufficient for PCs or PICCs [24]. Concerning anticoagulation prophylactic treatment, published results are conflicting. Older trials showed no reduction of catheter-related thrombosis, but a recent Cochrane review reported a reduction of symptomatic thrombosis with prophylactic heparin versus no heparin (RR=0.48, 95 % CI [0.27– 0.86]) [25–29]. It should be noted that in these studies, only few patients had a PICC as the central venous device. Considering the high rate of thrombosis with PICC in our study, anticoagulation prophylaxis remains a real issue. Some hypotheses can be postulated to explain the higher rate of complications in the PICC group. PICCs are inserted in smaller-diameter veins, and their catheters are longer compared to PCs. Thus, this prolonged contact between the PICC and the vascular wall might result in vascular endothelium damage and in a reduction in blood flow. These factors might participate in the development of vascular clots [21, 30]. Furthermore, some studies showed that several insertion attempts favored thrombotic complications causing endothelium damage [10, 31]. Contrary to PCs, the extremity of a PICC remains outside of the body and can more easily provide access for germs on the outer edge of the catheter. Weekly maintenance prevents cutaneous infectious complications from the insertion site, but repeated handling and flushes may favor bacteremia. Indeed, Yap et al. showed that dedicated nurses and nurse education on aftercare led to a decrease in the rate of PICC complications in their institution [32]. In our study, all relevant nurses received adequate education in PICC maintenance.

Conclusion In univariate and multivariate analyses, having a PICC instead of a PC as the central venous device was the only predictive factor of device-related major and minor complications in the administration of adjuvant chemotherapy for patients with breast cancer. This result requires confirmation with a prospective randomized study and an evaluation of patient quality of life. The NCT02095743 ongoing trial started in our institution will help address this important issue. If our results are confirmed by this prospective study, the management of

Support Care Cancer

adjuvant chemotherapy administration in breast cancer might be modified. Acknowledgments All the authors are employees of the Centre Henri Becquerel (Regional Center for the Fight against Cancer, Rouen, France), and they want to thank in particular the Department of AnesthesiologySurgery and the Department of Medical Oncology for their support. All the funds were provided by the Department of Medical Oncology. Conflict of interest The authors declare that they have no conflict of interest. We have full control of all primary data, which can be reviewed by the Journal on request.

15.

16.

17.

18.

References 19. 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J et al (2013) Cancer incidence and mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer 1990 49(6):1374–1403 Early Breast Cancer Trialists’ Collaborative Group (EBCTCG), Peto R, Davies C, Godwin J et al (2012) Comparisons between different polychemotherapy regimens for early breast cancer: meta-analyses of long-term outcome among 100,000 women in 123 randomised trials. Lancet 79:432–44 Roché H, Fumoleau P, Spielmann M et al (2006) Sequential adjuvant epirubicin-based and docetaxel chemotherapy for nodepositive breast cancer patients: the FNCLCC PACS 01 Trial. J Clin Oncol 24:5664–5671 Ng PK, Ault MJ, Maldonado LS (1996) Peripherally inserted central catheters in the intensive care unit. J Intensive Care Med 11:49– 54 Thiagarajan RR, Ramamoorthy C, Gettmann T, Bratton SL (1997) Survey of the use of peripherally inserted central venous catheters in children. Pediatrics 99:E4 Bottino J, McCredie KB, Groschel DHM, Lawson M (1979) Longterm intravenous therapy with peripherally inserted silicone elastomer central venous catheters in patients with malignant diseases. Cancer 43:1937–1943 Strahilevitz J, Lossos IS, Verstandig A et al (2001) Vascular access via peripherally inserted central venous catheters (PICCs): experience in 40 patients with acute myeloid leukemia at a single institute. Leuk Lymphoma 40:365–371 Walshe LJ, Malak SF, Eagan J, Sepkowitz KA (2002) Complication rates among cancer patients with peripherally inserted central catheters. J Clin Oncol 20:3276–3281 Chopra V, Anand S, Hickner A et al (2013) Risk of venous thromboembolism associated with peripherally inserted central catheters: a systematic review and meta-analysis. Lancet 382:311–325 Saber W, Moua T, Williams EC et al (2011) Risk factors for catheter-related thrombosis (CRT) in cancer patients: a patientlevel data (IPD) meta-analysis of clinical trials and prospective studies. J Thromb Haemost 9:312–319 Cheong K, Perry D, Karapetis C, Koczwara B (2004) High rate of complications associated with peripherally inserted central venous catheters in patients with solid tumours. Intern Med J 34:234–238 Ye X, Wong SW, Zhang J, Moo, IH et al (2014) Catheter-related upper limb venous thrombosis in a tertiary hospital setting. ANZ J Surg Kim HJ, Yun J, Kim HJ et al (2010) Safety and effectiveness of central venous catheterization in patients with cancer: prospective observational study. J Korean Med Sci 25:1748 Patel GS, Jain K, Kumar R et al (2014) Comparison of peripherally inserted central venous catheters (PICC) versus subcutaneously

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30. 31.

32.

implanted port-chamber catheters by complication and cost for patients receiving chemotherapy for non-haematological malignancies. Support Care Cancer 22:121–128 Schiffer CA, Mangu PB, Wade JC et al (2013) Central venous catheter care for the patient with cancer: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol 31: 1357–1370 Khavanin N, Gart MS, Berry T et al (2013) Sentinel lymph node biopsy versus axillary lymphadenectomy in patients treated with lumpectomy: an analysis of short-term outcomes. Ann Surg Oncol 21:74–80 Bafford A, Gadd M, Gu X et al (2010) Diminishing morbidity with the increased use of sentinel node biopsy in breast carcinoma. Am J Surg 200:374–377 Chew HK, Wun T, Harvey D et al (2006) Incidence of venous thromboembolism and its effect on survival among patients with common cancers. Arch Intern Med 166:458–464 Baron JA, Gridley G, Weiderpass E et al (1998) Venous thromboembolism and cancer. Lancet 351:1077–1080 Petterson TM, Marks RS, Ashrani AA et al (2015) Risk of sitespecific cancer in incident venous thromboembolism: a populationbased study. Thromb Res 135:472–478 Xing L, Adhikari VP, Liu H et al (2012) Diagnosis prevention and treatment for PICC-related upper extremity deep vein thrombosis in breast cancer patients. Asia Pac J Clin Oncol 8:e12–e16 Hanna H, Benjamin R, Chatzinikolaou I et al (2004) Long-term silicone central venous catheters impregnated with minocycline and rifampin decrease rates of catheter-related bloodstream infection in cancer patients: a prospective randomized clinical trial. J Clin Oncol 22:3163–3171 Darouiche RO, Berger DH, Khardori N et al (2005) Comparison of antimicrobial impregnation with tunneling of long-term central venous catheters: a randomized controlled trial. Ann Surg 242:193– 200 Van de Wetering MD, van Woensel JB (1996) Prophylactic antibiotics for preventing early central venous catheter Gram positive infections in oncology patients. Cochrane Database of Systematic Reviews Couban S, Goodyear M, Burnell M et al (2005) Randomized Placebo-Controlled Study of Low-Dose Warfarin for the Prevention of Central Venous Catheter–Associated Thrombosis in Patients With Cancer. J Clin Oncol 23:4063–4069 Verso M, Agnelli G, Bertoglio S et al (2005) Enoxaparin for the prevention of venous thromboembolism associated with central vein catheter: a double-blind, placebo-controlled, randomized study in cancer patients. J Clin Oncol 23:4057–4062 Young AM, Billingham LJ, Begum G et al (2009) Warfarin thromboprophylaxis in cancer patients with central venous catheters (WARP): an open-label randomised trial. Lancet 373:567–574 Karthaus M, Kretzschmar A, Kröning H et al (2006) Dalteparin for prevention of catheter-related complications in cancer patients with central venous catheters: final results of a double-blind, placebocontrolled phase III trial. Ann Oncol 17:289–296 Akl EA, Ramly EP, Kahale LA et al (2014) Anticoagulation for people with cancer and central venous catheters. Cochrane Database Syst Rev 10:CD006468 Caine GJ, Stonelake PS, Rea D, Lip GYH (2003) Coagulopathic complications in breast cancer. Cancer 98:1578–1586 Lee AYY, Levine MN, Butler G et al (2006) Incidence, risk factors, and outcomes of catheter-related thrombosis in adult patients with cancer. J Clin Oncol 24:1404–1408 Yap Y-S, Karapetis C, Lerose S et al (2006) Reducing the risk of peripherally inserted central catheter line complications in the oncology setting. Eur J Cancer Care 15:342–347