The CCSG prospective study of venous access devices: An analysis of insertions and causes for removal

The CCSG Prospective Study of Venous Access Devices: An Analysis of Insertions and Causes for Removal By Eugene S. Wiener, Patricia McGuire, Donna L. Betcher, Anneliese

Charles J.H. Stolar, R. Hampton

L. Sitarz, Jonathan D. Buckley, and G. Denman Hammond

Rich,

Vincent

C. Albo, Arthur

R. Ablin,

Mark D. Krailo, Connie Versteeg,

Pittsburgh, Pennsylvania; Denver, Colorado; New York, New York; Minneapolis, Minnesota; San Francisco, California; Rochester, Minnesota; Los Angeles, California l This is an interval analysis of the 2-year prospective multicenter Childrens Cancer Study Group study of 1,141 chronic venous access devices in 1,019 children with cancer. Device type was external catheter (EC) 72%, totally implantable (TID) 28%, and did not differ for diagnosis or age except more double-lumen devices in bone marrow transplant protocols (77%) and more TlDs in children less than 1 year old (17.7%). Insertion characteristics evaluated in 1,078 (95%) were: operating room placement 99%; general anesthesia 98%; cutdown 67%; percutaneous 33%; atrial position 50%, caval position 50%; and perioperative antibiotics 48%. Vein entry was the external jugular 33%. internal jugular 22%. subclavian 35%. cephalic 7%. and saphenous 3%. Insertion was difficult or very difficult in only 10% and operative complications occurred in only 0.7%. Degree of difficulty bore no relationship to device type or patient age. The reasons for removal in 736 devices (67%) were due to complications in 39%. of which infections were the most frequent. There was some variance between centers ranging from 8.5% to 31% for infection; 2.8% to 24% for dislodgment; and 0% to 13% for occlusion. ECs had a higher risk of dislodgment; elective removals were more frequent in TIDs; there was no difference in infection as a cause for removal between ECs and TIDs. Dislodgment was associated with the shortest distance of the cuff to the skin exit (mean, 4 cm): 5 2 cm, 49%; >2 cm, 28% (P = .009) and occurred most frequently in the younger patient (18.9%. 0 to 1 years; 0.5%, > 8 years). Copyright (~17992 by W.B. Saunders Company

INDEX WORDS: Central venous catheters; devices; implantable ports.

venous access

C (VADs) have gained widespread ENTRALLY

PLACED

venous access devices use and acceptance in the management of children with cancer and leukemia in whom frequent and/or prolonged access is required for chemotherapeutic agents, pain relief, intravenous fluids, antibiotics, blood products, total parenteral nutrition, and blood sampling. Patients and parents frequently request these devices when they see the ease with which other children are managed. Despite this widespread use there is little information concerning choice of device, methods of insertion, management, and complications. The Childrens Cancer Study Group (CCSG) conducted a prospective study (CCG S-31) of VADs to gain this information. This report is limited to the analysis of the insertion characteristics and causes for removal generated from this study. Other data will be reported separately. JournatofPedtatric Surgery, Vol.27, No 2 (February), 1992: pp 155-164

MATERIALS

AND METHODS

Eligibility requirements were: age 21 years or less, a diagnosis of cancer, insertion in a CCSG institution, and no long-term central venous device in the previous 6 months. CCSG investigators were encouraged to register all inserted devices. Each patient and newly placed device was registered at the time of insertion at the CCSG statistical office and assigned a study number, which was also used for any subsequent device placed in that patient during the study period. The CCG S-31 study was opened for patient entry on August 20, 1986, closed to new patient entry, except children 0 to 1 year of age, on May 31, 1988, and to all patients on December 15, 1988. The protocol was approved by the National Cancer Institute Clinical Trials Evaluation Branch and by each participating hospital’s Institutional Review Board. Informed consent was obtained before study entry and for each device insertion. Registration data came from the CCSG registration system used for all CCSG studies. Three data collection forms were specifically designed for this study and included the information tabulated in Table 1. If more than one device was placed in the same patient, each was entered and evaluated as a separate event. Only registration, insertion, and removal data were addressed in this report. VADs were designated as follows: external catheters (ECs), totally implantable devices (TIDs), double-lumen (DL), singlelumen (SL), small-lumen ([SmL] 2.7F to 4.4F), standard lumen ([StL] 6F to 7F), and large lumen ([LgL] 9F to 12F). The study committee established guidelines for device insertion, care, and usage, and for complication management. In addition each institution established its own device guidelines that were forwarded for committee review.

Statistical

Analysis

The significance levels for cross-tabulated data were calculated using the x’ test of the independence of traits across populations. When expected cell frequencies were less than 10, exact methods were used to calculate the significance level. When multiple

From the Children’s Hospital of Pittsburgh. Pittsbqh, PA; Children’s Hospital of Denver, Denver, CO; Babies Hospital, New York, NY; Minneapolis Children S Hospital, Minneapolis, MN; Universitv of California, San Francisco, CA; Mayo Clinic. Rochester. MN; Llniversity of Southern California School of Medicine, Los Angeles, CA. Supported in part by the Division of Cancer Treatment, National Cancer institute, National Institutes of Health, Department of Health and Human Services. Contributing Childrens Cancer Study Group investigators, institutions, and grant numbers are given in the Appendix. Presented at the 22nd Annual Meeting of the American Pediatric SurgicalAssociation, Lake Buena Vista, Florida, Mav 15-18, 1991. Address reprint requests to The Childrens Cancer Study Group, 440 E Huntingdon Dr, Suite 300, PO B0.x 60012, Arcadia, CA 910666012. Copyright Q 1992 by W B. Saunders Companv 0022-346819212702-0004$03.00!0 155

WIENER ET AL

156

Table 1. Data Capture Forms CCG S-31

to 7 devices. Six institutions entered 702 devices, 66% of the total study. Registration data are complete for all entered devices. At the time of this analysis, surgical check sheets have been completed for 1,078 devices (94%) and removal forms have been completed for 736 of the 766 (96%) removed devices. A comparison of the age distribution of patients registered on CCG S-31 to that of the tota CCSG patients registered during the same time period is shown in Table 2. Despite the attempt to increase the accrual of children aged 0 to 1 year, the number of children with registered VADs in that age group (17.7%) was less than that of other ages. A comparison of the diagnoses in the study children and in the overall CCSG is shown in Table 3. The most common oncologic diagnosis was leukemia (43%). Patients with bone tumors and neuroblastomas had the highest ratio of insertions (35%), whereas patients with Wilms’ tumors had the lowest insertion ratio (15%).

Surgical check sheet (completed by surgeon with operative note) Date of insertion Device type Method of insertion Indication for insertion Anesthetic used Difficulty of insertion according to grade shown below* Vein entry Tip position Tunnel length Cuff position Perioperative antibiotics Complication form (completed by nurse for each complication) Infection type, site, bacteriology, and treatment Catheter occlusion or vein thrombosis and treatment Catheter dislodgment Unable to withdraw blood Use and removal form (completed by nurse and/or data manager) Date of removal Cause for removal Device usage Device care Specify complications during the “life” of the device *Insertion

Device Type

difficulty: grading system. I. No difficulty: Catheter in-

serted in proper position at first attempt. Minor repositioning required to achieve final tip position. Insertion time

<45

minutes. II. Mild

difficulty: Several catheter manipulations required to achieve proper position. Insertion time < 1 hour. Ill. Difficult: Multiple manipulations required to achieve proper position. Only one entry site required. Insertion time

> 1 hour. IV. Very difficult: Initial access

site not

successful. One or more additional sites required. Insertion time > 1 hour. V. Complicated: Any or several of the following occur: never able to achieve central position, excessive

bleeding,

pneumothorax,

or

other significant operative or anesthetic complications.

adjustments for several factors were required, the multinomial logistic model was used.’ For all calculations, a P value of < 0.05 was considered significant. The number of catheter days contributed by each device was calculated as the number of days from insertion to removal or last patient contact, if the device was in place at the time of last contact.

Insertion information is available for 1,078 devices (Table 4). SL ECs were 44% of the total, SmL ll.l%, StL 20.8%, and LgL 12%. DL devices were 24.3% and TIDs were 26.9% of the total. Device type was not specified in 4.9%. Thus, ECs were chosen for 735 (72%) insertions and TIDs for 290 (28%). SmL ECs were used most commonly in children 3 years or less (79/120; 66%). DLs were used most frequently in children with leukemia, non-Hodgkin’s lymphoma, and neuroblastoma, who were potential bone marrow transplantation candidates (201/262; 77%). TIDs were used with the least frequency in children ages r3 years (65/343; 19%) and most frequently in children > 15 years (591124; 48%). No other patient demographic or diagnosis had any significant relationship to choice of device.

RESULTS

There were 1,141 devices inserted in 1,019 children from 25 CCSG institutions. Seventeen institutions entered at least 15 devices each; 8 contributed from 1

Insertion Characteristics

VAD insertion was performed in the operating room in 99% of cases, under general anesthesia in

Table 2. Age Distribution Age W

CCG S-31 Patients

EC

TID

All CCSGPatients*

S-31/Alit

1

113

82%

18%*

639

18%

2 to 3

274

80%

20%

a73

31%

4 to 7

268

77%

23%

1,059

25%

8tolO

122

68%

32%

506

24%

227

63%

37%

787

29%

137

52%

48%

427

32%

1,141

72%

28%

4,291

27%

oto

11

to 15

>I5 Total

Abbreviations: EC, external catheter; TID, totally implantable device. *All registered cancer patients in CCSG hospitals during study period. tPercent study patients compared with total CCSG in each age group. SP = ,013 patients I 1 year old compared with patients > 1 year old.

157

CCSG VENOUS ACCESS STUDY

Table 3. Common Oncologic Diagnoses Diagnostic Groups

CCSG Patients With VADt

Total CCSG Patients*

Leukemias

CCG S-31 Patients

Mean Age lyrl

6.9

1,850

27%

491 (43%)

Lymphomas

466

26%

121 (11%)

9.8

Brain tumors

481

24%

117 (10%)

7.1

Neuroblastoma

288

34%

99 (9%)

3.5

Sarcomas

394

22%

86 (8%)

7.3

Bone tumors

213

35%

74 (7%)

13.5

Nephroblastoma

444

15%

67 (6%)

3.3

Total

1,055

4,136

*All CCSG patients whether or not on study. tPercent of all CCSG patients entered onto CCG S-31 with VADs.

98%, by cut down in 66.7%, and percutaneously in 33.3% of cases. There was a trend toward more frequent percutaneous procedures as the patients became older (I 1 year, 24%; > 15 years, 40%). Only 7% of all percutaneous insertions were performed in children aged I 1 year. The most common veins of entry were: subclavian 34.8%, external jugular 33.4%, and internal jugular 22.4%. The cephalic and saphenous veins were used infrequently (7% and 3%, respectively). The right side was used in 70%. The tip of the catheter was positioned equally in the right atrium or in the superior vena cava. Perioperative antibiotics were used for 47.5% of the insertions. For ECs the average tunnel length was 11.2 cm and the average distance from the skin exit to the Dacron cuff was 4.1 cm. The insertion procedure was not difficult in 822 (76.9%) and only mildly difficult in 140 (13%). There was no significant difference in the difficulty of insertion according to device type, cutdown or percutaneous technique, and/or age of the patient. There were 8 (0.7%) complications associated with device insertions, of which 6 had cutdowns, 4 had DL ECs, and 1 was in a child I 3. Four were noncentral catheter tip positions, only one of which led to early VAD replacement. The other 4 were complications of the procedure: bleeding in 3, none of whom had thromboTable 4. Device Longevity NO.

Type of VAD

External catheter

735 (72%)

DD

DDiVAD

369,422

502.61 493.28

Single lumen Small

120

59,193

Standard

224

128,729

574.68

Large

129

57,716

447.41

Double lumen Totally implantable

262

123,784

472.46

290 (28%)

189,495

635.43 579.34

Other

53

30,705

Not available

63

5,919

1,141

595,541

Total Abbreviations:

521.95

VAD, venous access device; DD, device days; DD/

VAD, mean no. of days per device.

cytopenia, and postanesthetic aspiration in 1. None of the VAD placements led to patient deaths (Table 5). Causes for Device Removal

At the time of this analysis, 595,541 device days have accrued and 766 (67%) devices have been removed. TIDs accounted for 189,495 days or 653 days per device and ECs have accrued 369,422 days or 502 days per device (Table 4). Data on causes for removal are evaluable for 736 removed devices (Table 6). The most common causes for removal were not related to device complications: elective at end of therapy in 295 (40.1%) and death (noncatheter-related) with device in place in 157 (21.3%). For this report both of these categories combined were designated “elective.” Infection accounted for 47% of the 284 removals caused by device complications. In only two patients was death related to the presence of the device; both children died of catheter-related sepsis. A comparison of device types showed significantly more TID elective removals (73% 1’ 58%) and EC dislodgments (13% v 4%) in the entire cohort. These differences were not noted in the subset with leukemia. There were significantly more dislodgment removals for SmL ECs (22%) and StL ECs (16.8%) than the other ECs (8.6%). LgL and DL catheters had similar rates of elective removals (67% and 72%, respectively). Both had low rates of complications leading to removal when compared with all devices or with TIDs. More patients died with DLs in place (32%) than with the other VAD types. There was no difference in the percentage of ECs and TIDs removed due to infection. Dislodgment was more frequent when the cuff was closest to the skin exit site: 2 2 cm, 49%; > 2 cm, 28% (P = .009). This was confirmed in the subset with leukemia (P = .03). Dislodgment occurred most frequently in the younger age patient (0 to 1 years, 18.9%; 2 to 3 years, 12%; 4 to 7 years, 0.75%; and > 8 years, 0.5%). Tunnel length and cuff position had no

158

WIENER ET AL

Table 5. Difficulty of Insertion According to Device Type and Patient Age Age Iv) Device

ECst

TIDS

Grade*

2-3

<1

Mild

169 (89)

4-10

a3 (98)

211

Totals

232 (89)

171 (88)

656 (89.5)

Difficult

20 (11)

2 (2)

25 (10)

21 (11)

68 (9.5)

Complex5 Mild

0 (0) 37 (86)

0 (0) 20 (95)

4 (I)* 79 (91)

3 UP 121 (90)

7 (1) 257 (90) 27 (9.5)

Difficult

5 (12)

1 (5)

8 (9)

13 (10)

ComplexS

1 (2)

0 (0)

0 (0)

0 (0)

0.71

0.89

0.84

0.68

P value

1 (0.5)

NOTE. Numbers in parentheses indicate percents. Abbreviations: ECs, external catheters; TIDs, totally implantable devices. *See Table 1 for definitions. Mild, grades I and II; difficult, grades III and IV; complex, grade V. tThere was no difference in difficulty of insertions among the EC types. *Four were double-lumen ECs; P = .I 1. §Four noncentral tip locations (1 lasted 13 days, 2 lasted 6 months, 1 lasted 2.5 years); two arterial punctures (1 hematoma, 1 intrathoracic bleed into chest tube); one internal jugular tear with 100 mL blood loss; one postanesthetic aspiration required temporary reintubation.

significant effect on infection-related removals in the entire group or the subset with leukemia. Six CCSG institutions were responsible for 702 (66%) of entered devices and 526 (69%) of removals. In these six, the mean removal percentage was 75% (range, 66% to 88%). The causes for removal were: elective (mean, 63%; range, 50% to 88%), infection (mean, 21%; range, 8.5% to 31%), dislodgment (mean, 11%; range, 2.8% to 24%), and occlusion (mean, 4%; range, 0% to 13%). DISCUSSION

Broviac et al* first reported the use of silicone elastomer right atria1 catheters specially designed for long-term central venous access. Hickman et al3 later modified these catheters for use in bone marrow transplant recipients and cancer patients. Subsequently, TIDs attached to silicone or similar catheters were developed to allow for chronic intermittent central venous access without some of the disadvantages of external catheters.“6

The present study was performed to analyze the pattern of use and complications associated with chronic VADs in children with malignancies. The prospective, multiinstitutional design enabled a large number of devices to be evaluated in a similar fashion over a short time and provided a large data base. The 1,141 devices studied prospectively are significantly more than reported in any other childhood and most adult series,“” and allow analysis of variables not previously evaluated. However, this report does not address rates of complications, only those complications associated with device removal. The cohort of evaluated patients represents only a portion of all the children with malignancies registered in CCSG institutions during the study period. The number or outcome of devices placed in the non-CCG S-31 patients in unknown. However, there is no reason to suspect selection bias relative to patient diagnosis or device management that would have influenced the validity of data analysis or conclusions. The six institutions that contributed two thirds

Table 6. Device Type Versus Reason for Removal Complication

Elective Type of Device External catheters

Elective 213 (37)

Death (NR)

Infection

Occlusion

Dislodgment

Other

119 (21)

107 (19)

22 (4)

74 (13)

36

Single lumen Small

26

13

19

3

19 (22)X

6

Standard

76

14

28

7

27 (17)’

9

Large

44

23

19

0

a

6

56

60 (32)”

36

10

17

11

82 (50)*

38 (23)

27 (16)

Double lumen Totally implantable Other Total

11 295 (40)

9 157 (211

5 134 (18)

5 (3)

7 (4)*

6

2

3

4

27 (4)

ai

(II)

42 (6)

NOTE. Data given in numbers of patients. Numbers in parentheses indicate percents. Total elective removal: 452 (61%); total removal due to complication: 284 (39%). Abbreviation: NR, nondevice related. *P < .05.

CCSG VENOUS ACCESS STUDY

of the study devices consecutively entered virtually all of their VADs placed during the study period. Analysis of this smaller but more representative cohort confirmed the findings in the total study population, although some variation between centers was noted. There was little variation in results when evaluated by age and diagnosis groups. It was somewhat surprising, considering the difficulty in infant venous access, that there were proportionately less VADs registered in infants 2 1 year of age. This may reflect the prevailing concern of some investigators of more VAD complications in this age group. However, VAD insertion was not more difficult or more complicated in any age group or for any device type. Age alone should not be a consideration in determining whether a child should have a VAD inserted. VADs were used in Wilms’ tumor patients less frequently than in the other diagnosis groups. This probably reflects the shorter length of administration and lesser intensity of chemotherapy required in nephroblastoma, factors that should be considered when determining if a device is needed. All VAD insertions in this study were by surgeons and virtually all in the operating room using general anesthesia. The facility of insertion, which was not difficult in more than 90%, and the extremely low insertion complication rate of 0.7%, which compares favorably with other reports,7.1”‘4 appears to justify this approach. Two thirds of catheters were inserted by cut down, most often into the external jugular vein, although there was some institutional and individual surgeon variation and preference. This approach in childrer? is contrary to the adult experience in which a percutaneous approach to the subclavian system is most commonly used.“,” The percutaneous approach has been advocated in children as well.‘R,‘6Very few catheters were inserted through inferior caval routes.27 The preponderance of the right-sided approach probably reflects recommendations in the literature2R~‘” and the need to keep the left chest clear for echocardiography. The majority of VADs in this study (61%) were removed electively at the end of therapy or were in place at time of patient death. In only two cases was the death associated with a catheter complication (sepsis). The average device longevity in these frequently sick, septic, poorly nourished, multiply accessed patients was 522 days. This compares favorably to the average time on study for all CCSG protocols of 547 days. More elective removals were reported in the TID group (73% v 58%) and their average life span was longer (653 days). Despite the preponderance of elective removals, it

159

is nonetheless disturbing that 39% of removals were caused by complications. Stanislav et al’” removed 30 of 115 (26%) devices due to complications. Mirro et al’ reported a 21% complication-related removal rate. Cameron” had a 24% complication-removal rate. However, it should be noted that the length of follow-up in CCG S-31 is longer than in these other reports. With 67% of all devices having been removed at the time of this analysis, mean follow-up was 541.4 -+ 14.8 days. Thus, this higher rate of complication-related removals may be attributed to the length of follow-up. Infection was the most frequent complication resulting in device removal. This has been the universal experience. King et al” reported a sepsis incidence of 22% in childhood oncology patients. In 74% of their infected patients, antibiotic administration avoided device removal, an experience echoed in other reports.y.“2~34Several reports7.‘3.35-37 have noted a decreased rate of infection for TIDs compared with ECs. The S-31 data did not show a difference in infection-related removals between ECs and TIDs. The large number of dislodgment-related removals of ECs may have resulted in their being at less risk to develop subsequent infections. Broviac’s contribution to catheter design was to add a molded hub and a Dacron cuff to a silicone catheter.’ Although not well established by clinical trials, a tunnel from the vein entry to the skin exit guarded by a cuff may theoretically decrease ascending infection.3x-4”The S-31 data showed no infectionprevention benefit to tunnel length or to cuff position. Shorter tunnels and smaller distances of the cuff from the exit are interrelated. If shorter cuff distances cause more dislodgments, those ECs with longer tunnels were in place, and at risk, longer to develop subsequent infections. This may have masked any infection-protecting benefit of a tunnel. Children with leukemia were analyzed separately in an attempt to determine whether the tunnel length analysis was influenced by the more intensive therapy received in some of the solid tumors which occur primarily in older and larger children. Leukemia receives relatively homogeneous therapy across all ages and sizes. The lack of benefit of long tunnels was confirmed. Catheter displacement from a central position or complete dislodgment was a major problem that led to 13% of EC removals. Dislodgment caused removal most frequently in children r3 years of age and of the smaller lumen ECs generally placed in this younger age group. Dislodgment of the catheter requiring VAD removal occurred even with TIDs but less than with ECs. Mirro et al’ and Cameron”’ each reported a 10% dislodgment incidence in children,

160

whereas Petersen et a141reported that 3% of catheters were accidently pulled out by the adult patient. Although tissue ingrowth into the Dacron cuff presumably prevents accidental dislodgment, this occurs over a period of time and is influenced by the patient’s nutritional status, immunocompetence, treatment (steroid effect), and presence of local infection. This study demonstrated a definite influence of cuff position on dislodgment. Catheters with cuffs within 2 cm of the skin exit had a significantly greater chance of dislodgment. In order to prevent early dislodgment, several authors31,42recommend placing the cuff 5 to 8 cm from the exit site and/or also anchoring the cuff to the subcutaneous tissues with a suture to stabilize the catheter during the period of tissue ingrowth.4’,43 The tunnel should be long enough so that the cuff can be positioned at least 2 cm from the exit site, whenever possible. The higher dislodgment rate in younger children may warrant using Vazquez’s technique in this age group. Occlusion of the catheter was an uncommon cause for removal of either ECs or TIDs. This reflects the value of heparin flushes in thrombus prevention and the ability to salvage occluded devices with urokinase, both of which were used in this and other studies.15.44-48 A variety of devices are currently available. The devices used in this study were chosen by oncologists, surgeons, oncology nurses, parents, and patients. It was not unusual for a patient to have several different types of devices. ECs were used in 72% of cases. TIDs became more popular as the study progressed. Review of insertion forms showed that some institutions clearly favored external catheters while others preferentially inserted totally implantable devices. Some institutions preferred multilumen catheters, especially when bone marrow transplantation was anticipated. The analysis of the difficulty of insertion according to EC type, ECs versus TIDs, and patient age (Table 5) showed no differences. The type of device had no influence on the difficulty of insertion nor the incidence of operative complications irrespective of patient age. Based on the results of this study and a review of the pertinent literature, device type can be selected in a rational manner: Lumen Size

The size of the catheter generally mirrored the size of the patient. There was no evidence of more difficult insertions or increased complication-related removals when VADs with larger lumens were used. There were not enough larger-lumen devices placed in young children to evaluate the risk according to

WIENER ET AL

age. In a study by Mirro et al,’ there was no difference in complications between StL and LgL catheters. However, catheter size analysis was not adjusted for patient age. Cameron3’ recommends standard size catheters for all oncology patients. There are no data to support selection criteria for lumen size. However, data available from this and other studies indicate that all sizes of catheters can be safely inserted when catheter size is adjusted for patient size and age. Single-Lumen Versus Multilumen

Multilumen devices have become increasingly popular especially in the care of critically ill patients. In a prospective evaluation of SL catheters versus triplelumen catheters (TLC) in adults,49 significantly more TLCs were removed because of infection (P = .029). Ulz et alSo described 88 adults with DL catheters; complications resulted in removal of 25%. Petersen et a141recommended that DL catheters be inserted if bone marrow transplantation is anticipated; however, none of 7 children with a SL device needed an extra catheter. Cameron31 concluded that the extra maintenance required did not seem justified by the infrequent benefit of DL catheters. In CCG S-31, DLs were not associated with increased complicationrelated insertions or removals. Although 4 of the 8 complicated insertions occurred with DL catheters, this did not achieve significance (P = .114). However, due to the type of patient and diagnosis in which they were used, more of these catheters were in place at the time of patient death. This may have allowed fewer of these devices remaining in place to develop subsequent complications. There were a few DL implantable devices placed in this study, but not enough to analyze separately. Multilumen catheters are recommended for potential bone marrow recipients and for other children in whom multiaccess is anticipated. ECs Versus TIDs

TIDs can be inserted safely and without undo difficulty, even in infants. Complications appear to be less than with ECS.‘.~‘-~’ Ross et a1,13in a nonrandomized study in 90 children, inserted 50 TIDs with less infection, better patient acceptance, and lower total maintenance costs than 40 ECs. Stanislav et alI4 represent the present adult oncology preference for TIDs, which accounted for 62% of total devices in 115 adult cancer patients. The present report showed that significantly less TIDs were removed due to all complications, most specifically dislodgment in the younger patient.

161

CCSG VENOUS ACCESS STUDY

The longer life expectancy, ease of care, improved patient acceptance, and decreased complicationrelated removals with TIDs suggest that they are the preferred device except in situations in which continuous access or bone marrow transplantation is anticipated.

Participating

Safe placement of either ECs or TIDs is possible in all age groups and for all diagnoses. They can last the entire treatment course in most patients. Some unplanned removals are preventable by specific insertion techniques. Careful patient and device selection are essential.

Appendix Principal Investigators-Childrens Study Group

Institution Group Operations Office, University of Southern California Comprehensive Cancer Center (Los Angeles. CA)

University of Michigan Medical Center (Ann Arbor, MI) University of California Medical Center (San Francisco, CA) Rainbow Babies and Children’s Hospital (Cleveland, OH) Children’s Hospital National Medical Center (Washington, DC) Children’s Memorial Hospital (Chicago, IL) Children’s Hospital of Los Angeles (Los Angeles, CA) Children’s Hospital of Columbus (Columbus, OH) Babies Hospital (New York, NY) Children’s Hospital of Pittsburgh (Pittsburgh, PA) Vanderbuilt University School of Medicine (Nashville, TN) University of Minnesota Health Sciences Center (Minneapolis, MN) Children’s Hospital of Philadelphia (Philadelphia. PA) Memorial Sloan-Kettering Cancer Center (New York, NY) James Whitcomb Riley Hospital for Children (Indianapolis, IN) University of British Columbia (Vancouver, BC) Children’s Hospital Medical Center (Cincinnati, OH) Harbor/UCLA and Miller Children’s Medical Center (Torrance and Long Beach, CA) University of California Medical Center (Los Angeles, CA) University of Iowa Hospitals and Clinic (Iowa City, IA) Children’s Hospital of Denver (Denver, CO) Mayo Clinic (Rochester, MN) University of North Carolina (Chapel Hill, NC) University of Medicine and Dentistry of New Jersey (Camden, NJ) Children’s Mercy Hospital (Kansas City, MO) Wyler Children’s Hospital (Chicago, IL)

Cancer Grant

Investigator

No.

Denman Hammond, MD John Weiner, DrPH Harland Sather, PhD Mark Krailo, PhD Jonathan Buckley, MBBS, PhD Madeline Bauer, PhD Daniel Stram, PhD Raymond Hutchinson, MD

CA 13539

Authur

CA 17829

Ablin, MD

Susan Shurin, Gregory

CA 20320

MD

Reaman,

MD

CA 03888 CA 07431 CA 02649

Edward Baum, MD Jorge Ortega, MD Frederick

CA 02971

Ruymann,

MD

CA 03750

Sergio Piomelli, MD Vincent Albo. MD

CA 03526 CA 36015

John Lukens,

CA 26270

MD

William Woods,

MD

CA 07306

Anna Meadows,

MD

CA 11796

Peter Steinherz,

MD

CA 42764

Robert

Weetman,

MD

Paul Rogers, MD Robert Wells, MD Jerry Finklestein, Stephen Raymond

CA 29013 CA 26126 MD

Feig. MD Tannous,

CA 13809

CA

14560

CA 27678 MD

David Tubergen, MD Gerald Gilchrist, MD Herbert Cooper, MD Milton Donaldson, MD Arnold Freeman, MD F. Leonard Johnson, MD

CA 29314 CA 28851 CA 28882

162

WIENER ET AL

REFERENCES 1. Bishop YMM, Feinberg SE, Holland PWM: Discrete Multivariate Analysis. Cambridge, MA, MIT Press, 1975 2. Broviac .I, Cole B, Scribner B: A silicone rubber atrial catheter for prolonged parenteral alimentation. Surg Gynecol Obstet 136:602-606, 1973 3. Hickman R, Buckner C, Clift R. et al: A modified right atria1 catheter for access to the venous system in marrow transplant recipients. Surg Gynecol Obstet 148871-875, 1979 4. Gyves J, Ensminger W, Niederhuber J, et al: Totally implanted system for intravenous chemotherapy in patients with cancer. Am J Med 73:841-845,1982 5. Niederhuber J, Ensminger W, Gyves J, et al: Totally implanted venous and arterial access system to replace external catheters in cancer treatment. Surgery 92:701-711,1982 6. Wright KC, Szwarc IA, Collins RD, et al: A new subcutaneous vascular access device: An experimental evaluation. Radiology 146:222-22331983 7. Mirro J Jr, Rao BN, Kumar M, et al: A comparison of Hickman/Broviac catheters and subcutaneous ports in children with cancer. J Pediatr Surg 25:120-124,199O 8. Doran I? Care of the Hickman catheter in children. Nurs Clin North Am 18:579-583,1983 9. Shapiro E, Wald E, Nelson K, et al: Broviac catheter-related bacteremia in oncology patients. Am J Dis Child 136:679-681.1982 10. Thomas J, MacArthur R, Pierce G, et al: Broviac catheters: Indications and results. Am J Surg 140:791-796, 1980 11. Wade J, Newman K, Schimpff S, et al: Two methods for improved venous access in acute leukemia patients. JAMA 246: 140144,1981 12. King DR, Komer M, Hoffman J, et al: Broviac catheter sepsis: The natural history of an iatrogenic infection. J Pediatr Surg 20:728-733,1985 13. Ross M, Haase GM, Poole MA, et al: Comparison of totally implanted reservoirs with external catheters as venous access devices in pediatric oncology patients. Surg Gynecol Obstet 167: 141,1988 14. Stanislav GV, Fitzgibbons RJ, Bailey RT, et al: Reliability of implantable central venous access devices in patients with cancer. Arch Surg 122:1280-1283,1987 15. Stokes DC, Rao B, Mirro J, et al: Early detection and simplified management of obstructed Hickman and Broviac catheters. J Pediatr Surg 24:257-262,1989 16. Davis SJ, Thompson JS, Edney JA: Insertion of Hickman catheters: A comparison of cutdown and percutaneous techniques. Am Surg 50:673-676,1984 17. Pietsch JB, Nagaraj HS, Groff DB: Simplified insertion of central venous catheter in infants. Surg Gynecol Obstet 158:91-92, 1984 18. Newman BM, Jewett TC Jr, Karp MC, et al: Percutaneous central venous catheterization in children: First line choice for venous access. J Pediatr Surg 21:685-688,1986 19. Aitken DR, Catalan0 R, Minton JP: Central venous access in oncology patients: The “Peel-Away” sheath for rapid insertion. J Surg Oncol22:81-83,1983 20. Pereyra R, Andrassy RJ, Mahour GH: Central venous cannulation in neonates. Surg Gynecol Obstet 151:253-254, 1980 21. Newsome HH Jr, Armstrong CW, Mayhall GC, et al: Mechanical complications from insertion of subclavian venous feeding catheters: Comparison of de novo percutaneous venipuncture to change of catheter over guidewire. J Parenter Enter Nutr 8:560-562, 1984 22. Slezak FA, Williams GB: Delayed pneumothorax: A compli-

cation of subclavian vein catheterization. J Parenter Enter Nutr 8:571-573,1984 23. Marino P: Delayed pneumothorax: A complication of subclavian vein catheterization. J Parenter Enter Nutr 9:232, 1985 (letter) 24. Kawamura R, Okabe M, Namikawa K: Subclavian vein puncture under ultrasonic guidance. J Parenter Enter Nutr 11:505506, 1987 25. Nicholson SC, Sweeney MF, Moore RA, et al: Comparison of internal and external jugular cannulation of the central circulation in the pediatric patient. Crit Care Med 13:747-749, 1984 26. Gauderer MWL, Stellato TA: Subclavian Broviac catheters in children-Technical considerations in 146 consecutive placements. J Pediatr Surg 20:402-405,1985 27. Kanter RK, Zimmerman JJ, Strauss RH, et al: Central venous catheter insertion by femoral vein: Safety and effectiveness for the pediatric patient. Pediatrics 77:842-847, 1986 28. Sullivan CA, Konefal SH Jr: Cardiac tamponade in a newborn: A complication of hyperalimentation. J Parenter Enter Nutr 11:319-321, 1987 29. Agarwal KC, Ali Kahn MA, Falla A, et al: Cardiac perforation from central venous catheters: Survival after cardiac tamponade in an infant, Pediatrics 73:333-338,1982 30. Tocino IM, Watanabe A: Impending catheter perforation of superior vena cava, radiographic recognition. AJR Am J Roentgenol 146:487-490, 1986 31. Cameron GS: Central venous catheters for children with malignant disease: Surgical issues. J Pediatr Surg 22:702-704, 1987 32. Rose SG, Pitsch RJ, Karrer FW, et al: Subclavian catheter infections. J Parenter Enter Nutr 12:511-512, 1988 33. Lazarus HM, Lowder JN, Anderson JM, et al: A prospective randomized trial of central venous catheter removal versus intravenous Amphotericin B in febrile neutropenic patients. J Parenter Enter Nutr 8:501-505,1984 34. Faubion WC, Wesley JR, Khalidi N, et al: Total parenteral nutrition catheter sepsis: Impact of the team approach. J Parenter Enter Nutr 10:642-645,1986 35. Opie J, Chan K, Rogers P, et al: Implantable intravenous access devices in children. BC Med J 29:274-276.1987 36. Saucy P: Experiences with the use of the Port-A-Cath in children. J Pediatr Surg 22:767-769, 1987 37. McGovern B, Solenberger R, Reed K: A totally implantable venous access system for long-term chemotherapy in children. J Pediafr Surg 20:725-727,1985 38. Holt RW, Heres E: Creation of a subcutaneous tunnel for Broviac and Hickman catheters. J Parenter Enter Nutr 9:225,1985 39. Klein MD, Coran AG: A technique for tunneling central venous catheters. J Parenter Enter Nutr 9:521-522,1985 40. Bothe A Jr, Piccione W, Ambrosino JJ, et al: Implantable venous access system. Am J Surg 147:565-568,1984 41. Petersen FB, Holo M, Hickman RO, et al: Fate of right atrial catheters inserted prior to arrival at a transplant center. J Parenter Enter Nutr 11:263-266,1987 42. Vazquez RM. Racenstein MJ: Amethod to prevent unintentional removal of a Hickman catheter. J Parenter Enter Nutr 11:509-510, 1987 43. Schaller R: Personal communication, April 1989 44. Hurtubise M, Bottino J, Lawson M, et al: Restoring patency of occluded central venous catheters. Arch Surg 115:212-213, 1980 45. Lawson M. Bottino J, Hurtubise M, et al: The use of urokinase to restore patency of occluded central venous catheters. Am J Intravenous Ther Clin Nutr, 1982 46. Delaplane D, Scott J. Riggs T, et al: Urokinase therapy for a catheter-related right atria1 thrombus. J Pediatr 100:149-152, 1982

163

CCSG VENOUS ACCESS STUDY

47. Glynn M. Langer B, Jeejeebhoy K: Therapy for thrombotic occlusion of long-term intravenous alimentation catheters. J Parenter Enter Nutr 4:387-390, 1980 48. DiCostanzo J, Sastre B, Chous R. et al: Experimental approach to prevention of catheter-related central venous thrombosis. J Parenter Enter Nutr 8293297, 1984

49. McCarthy MC. Shives JK, Robison RJ. et al: Prospective evaluation of single and triple lumen catheters in total parental nutrition. J Parenter Enter Nutr 11:259-262, 1987 50. Ulz L. Petersen FB, Ford R, et al: A prospective study of complications in Hickman right atrial catheters in marrow transplant patients. J Parenter Enter Nutr 1427-30. 1990

Discussion D.R. King (Columbus, OH): I think Dr Wiener and the committee should be congratulated for organizing this vast amount of data. I can’t agree with all of their conclusions, however, and I would like to ask a couple of questions. The TIDs seem to be favored because there was an increase in the number of elective removals. However, this included the patients who died with the catheters in place and there was an increased number of those in the TID patients. Does the statistical advantage hold up if only the truly elective removals are included and not those patients who died with this device in place? Second, the most common complication of the ECs was sepsis. Many studies have affirmed the fact that 75% of infected catheters can be cleared of infection with antibiotic therapy. Were all of these children treated with antibiotics prior to catheter removal? Third, there is a tremendous disparity in the sepsis rates reported at different institutions (varying from 8% up to 30%). This suggests an internal inconsistency in reporting that might invalidate the data. Were these interinstitutional differences significant from a statistical standpoint? Fourth, were the catheter dislodgements an early or late phenomenon? Were these children dislodging the catheters at a week postoperation because the surgeons were not appropriately putting fixation stitches at the skin exit site or were they being dislodged at the end of 6 months because of poor tissue ingrowth into the dacron cuff? And lastly from a personal standpoint, what has this study done in terms of changing your practice as far as management of these catheters is concerned? P.S. Liebert (Eastchester, NY): I know that this was not listed in the presentation as a complication of catheter insertion, but I just wonder how many times the catheter ruptured or needed to be repaired? I had an unusual situation recently in a 12-year-old patient who decided that he did not like the catheter, took a pair of scissors and severed the external portion. Was the incidence of catheter leak or rupture included in the study? R.M. Filler (Toronto, Ontario): I would just like to make one comment and ask a question. We have looked at our own experience which has extended

over many years. In comparing catheters and ports inserted for chemotherapy, we found there was a very significant difference in infection rate in that the rate for ports was about half that for external lines and both were very much less than the rate for lines that were used for total parenteral nutrition. Our statistical analysis was somewhat different than yours because we took into account the number of days a catheter was in place. Obviously catheters left in for longer periods have a greater chance of getting into trouble. The standards that we have used is how many infections per thousand catheter days. When you do that kind of statistic, you get a different result than in calculating the percent of catheters infected. I wonder if these data are available? S.J. Shochat (Stanford, CA): I didn’t see any cases in which the facial vein was used. Are those cases under the internal jugular? Did you have infiltration of the ports? Did you correlate the absolute neutrophil count (ANC) with infection? Several years ago, we presented a paper in which there was a definite correlation between a low ANC and infection at the time of insertion and shortly thereafter. M.P. LaQuaglia (New York, NY): I would echo Dr Filler’s statements about the time dependent analysis. We find that ports are better in terms of infection. One of the things that we have noticed is that the ports are harder to deal with when they have trouble. Are there any hard data as to whether vesicants are actually harder to give through ports than through external lines? E.S. Wiener (response): This study attempted three things. First, what happened at the time of insertion of these devices. Second, what complications occurred during the life of the devices. Third, what happened at the time of removal. Unfortunately, we do not have reliable data about complications that occurred during the life of the catheter. We do have a lot of data but the committee and the statisticians do not feel that they are reliable and should be reported at this time. I did not report rates of infection and I did not report catheter days. We have reliable data at the time of insertion and removal. Dr King, if only elective removals are counted the benefit of TIDs

164

does hold up. If the removal is done for one reason then it can’t occur for another reason. So, for example, if ECs have a lot of dislodgments then they don’t remain in place long enough to develop infections. Dislodgments did tend to occur early. These data can’t answer reliably your question about infection and salvage of the catheter. Many catheters were salvaged. There was a disparity among the six major institutions that contributed 66% of the data. However, there was not any significant difference in the elective removal rate. What differed was that some institutions have more infections while some had more occlusions as cause for removal. Part of that may be the way we looked at the data. We only chose one cause for removal; sometimes two causes were reported and we had to pick one. As far as what this

WIENER ET AL

has done to my practice, and I think we have seen this over the course of the study, and not necessarily because of the result of the study, more totally implantable devices are being inserted. Dr LaQuaglia, there is no hard evidence that vesicant therapy is worse or continuous infusion is worse for ports. Some institutions have put almost all TIDs in for every purpose while others are very much more selective. Dr Liebert, I can’t answer your question about rupture and repair for the reasons I have already mentioned. Dr Filler, the catheter rate per day that we have is better then anyone’s reported data, but I don’t believe it is reliable. Dr Shochat, the facial vein was included in the internal jugular vein sites. Absolute neutrophil count at the time of insertion did not affect infection causing removal.