Improved Methods for Venous Access: The Port-A ...

f

Improved Methods for Venous Access: The Port-A-Cath, A Totally Implanted Catheter System By Stephen Strum, J o n o k n McDermed, Alexander Korn, and Corrine Joseph We prospectively evaluated the performance and rate of long-term complications with the Port-A-Cath (PAC), a totally implanted vascular access system. Two catheter st& were evaluated, a small-bore (SB) catheter (0.51-mm diameter) and a large-bore (LB) catheter (1.02-rnm diameter), i n conjunction with the use of a strict catheter care protocol. The PAC performed well, and with both SB and LB systems, no significant extmvascltion, skin necrosis, hematoma, septum damage M leakage, or subcutaneous portal infections occurred ofter 7,240 dayr of implantation and 1,435 days of use. Complications with the PAC system consisted of catheter occlusion (seven patients, 21.5%) and one instance of possible catheter infection (3.1%). Occlusions were limited to patients im-

v

ASCULAR ACCESS PROBLEMS fiequently develop in cancer patients receiving chemotherapy. In a recent analysis of patient perceptions of their therapy. cancer patients rated the pain associated with the search for suitable veins to be one of the most distressing physiologic side effects of chemotherapy.' Venous initation from anticancer drugs and the need for repeated venipuncture for a period of months to years results in gradual sclerosis, thrombosis, and the destruction of available surface veins. In patients with poor peripheral veins, many oncologists recommend the placement of a centxal venous catheter for venous access to circumvent the high risk of extravasation from administration of vesicant chemotherapy. A relatively new hethd of venous access is being explored through the use of totally implanted vascular access ports. These systems reFrom thc Dcpartmenu of Medicine and Clinical Pharmacy. Uniursiiy cf Sourhcrn California. rht Norris Cancer Hospilal and Research I n t t i ~ rLos , Angcles; and tht Division o/Mrdical Oncology and Hcnntology. B r o m Medical Ccruer. Culw r Ciiy. Calif. Svbmincd March 25. 1985; arccpred Oct 18. IW. Address rcpritu requests to Srephen B. Strum. MD. 9808 Venice Blvd. I S O 3 . Culvcr Ciry. CA 90230. 0 1986 by American Socicry qf Clinical Oncology. 0732-lB)X/86/0404-0001IU.00~0

planted with the SB catheter (seven of 16, 43.8%), and five of the seven (71.4%) occurred i n patients receiving continuous infusion chemothempy and/or total p o k n t e m l nutrition. Patency of the PAC sysiern was maintained using a regular flushing schedule once every 30 days, a significant advantoge mp a d with the daily maintenance schedule required with externally placed venous catheten. The results of this study suggest that the PAC system can provide a safe and reliable method for venous acceu i n potients requiring intermittent or prolonged intravenous therapy. J Clin Onrol4:596403. O 1986 b y ~ m c n ' & Joricty of Clinical Oncology.

tain all of the advantages of external venous catheters with respect to central venous access for drug administration, total parenteral nutrition (TPN), or repeated phlebotomy. The subcutaneous location of the injection site is esthetically desired by some patients and allows uninhibited patient activity when the catheter is not in use. Maintenance requirements for implanted access ports are minimal, and because of .the lack of externally exposed parts and decreased catheter manipulation, the risk of catheter-related sepsis is thought to be I O W . ~ . ~ One such system is the Port-A-Cath (PAC) (Pharmacia Nu-Tech, Piscataway, NJ). It differs from other access ports in that a variety of Silastic catheters can be connected to the stainless steel injection portal to provide access for intravenous (IV), intraarterial, or intraperitoneal drug administration. Preliminary reports of clinical experience with the venous PAC have been p u b lished from major teaching institutions.'.' However, little information is available regarding rhe long-term performance and complication rate with the PAC used in the community setting. The purpose of this study was to present follow-up data on 32 patients who were implanted with the PAC venous access system and were treated in a community setting.

l a r o l of Cliniml O d o g y , V d 4, No 4 (April), 1986: pp 596603

PAC-AN

WROVED METHOD F O R VENOUS ACCESS Table 1.

Patient Demogmphio

Type of Catheter

Chomaairtk Total number of patients Sex (M:F) Agc (yr)

SB

LB

16

16 6: 10

3: 1 3 63.8 68 4148

Mccn M i

Ran~e

66.1 67 52-81

Primay m o l i i

Patients were provided w i t h written information describing the PAC implant system as well as a description of available external central venous catheten. The pmcntial advantages and didvantages of these access devica ~+rch m u g h l y disc u d . Patients were informed of the pmpov of the research study and the investigational status of rbc PAC system. Thirtytwo of 35 eligible patients a p e 4 to be cmdled into the study. The p~otocolwas approved by both the human research and h u m rights comminees. trhich constiarc the institutional review b o d a l Braman M a i d Center. Winen informed consent was obtained in dl pYients.

The PAC was purchrued from Ph.?nnxia Nu-Tech (Piscauway. NJ). It consisted d two components: ( I ) a stain& weel injection p o d that housed a higbdccuity. self-sealing silicone scpum and (2) i n m h a n g e a b k Sillstic catheters that were connected to the poctal with a lockkg ring (Fig I). The two cuheter styles studied were cmstructed of identical inen mavrid s and differed only in heir bore d-i.

Surgical Procedure MATERIALS AND METHODS This was an uncoclmlled clinical evaluation of the performance and m e d mplications encountered with h e use of the PAC venous mss pon. Two catheter styles were evaluared in the coursc of the srudy: a snull-bore (SB) catheter (0.51-rnm diameta) and a luge-bore (LB) catheter ( I .02-mm diamter). Betwan Jlnuuy and May 1983, all patients enrolled w i v e d the SB syszrn implant. All patients subxqucntly enrolled received the LB system implant. which had become available fa use after May 1983.

Patients The populiion d i e d cauisted of 32 cancer patients uho were candidates f a central venous catheter placement f a purpovs ofchcrnotknpy and/or TPN. Patient demographic dab is outlined in Table I. Sixteen patients received the SB system implant. md 16 patients received the LB system. nK PAC was implanted for TPN and cherochcnpy administration in 24 of 32 patients. continuous. lowdose chemochenpy in two patients and &or(-term intminent chemotherapy in six patients. Eleven patients nccivcd TPN for less than 30 days. whereas 13 odxn received TPN foc more than 30 days.

Undcr local anesthesia. the catheter ru intlpduced into the superior vena cava via either the internal jugular, cephalic. a subclavian venous mule with the aid of a guide wire. Ihe pmxima1 end of the tamer w n tunneled bcnuh Ihc skin into a subcutaneous pocker created in the r i g h a k h inhaclaviculu spacc and was connected to tbc portal wxth the laking ring (Fig 21. The system was filled w i l h heparinizrd saline and the p o d implanted and secured into p i t i o n to urdalying musck fascia using nonabsorbabk 2.0 Edubond s u m (Wlicoa. Summervilk. NJ) on an O S 4 nee& a the four nrmrr: loop sius. The incision was closed wifh %-Strips (3M. St Paul) and the surgical site covered with a sterile vansparrnr dressing. Catheter tip location was determined by chest before administ n h n of drugs. IV fluids. a TPN solutiom through the PAC.

Injection of the Subcutaneous Porral Rr prescribed protocol, flushing or drug administration could only be pcrfonned after receiving the in-servicing required by the tnanuf~turerand by thoY p e m n n d formally mined in the c u r and maintenance of the PAC. Hamhuhlng wifh a bactmcwld soap or scrub was rrqoired. and tbc use of sterile gloves and aseptic technique during the procedure was mandatory . In the majority of patients. rhe portal c o d be easily palpated blow the overlying skin. and Ihe portal v p u m could bc located Fig 1.

(A) -ic

diagram

of the PAC system showing i h component parts: (1) stainless steel portal housing with a highdensity, self-sealing silicone sep turn; (2) a catheter lock that mates the standard portal with various m t k t e r stylw and (3) a Silastic catheter that is cut to the desirud length at the time of surgical implantation. (6) Cou-sectional diagmrn depicting the direct method d PAC o c c w using a syringe and stmight Hubw needle.

PAC-AN

IMPROVED METHOD F O R VENOUS ACCESS

py were classifmi into the long-urn calegory. In l h e v p i c n u , r ! ~PAC was ~nuse faexlcndcd penodr. in some. for the entire lim thc system was implanted. Monitoring panmcters were &fined to assess the integrity and relative occumoce of complications with the PAC. Implanu t i m time uas &fined as the number of days the system was in p l r e . from the dale of implanution to the date complication(s) a dcath occumd. This parameter was used as a global assessment of PAC integrity. Time uud was defined as the tolal number of days the system was xrually in use and was used to evaluate catheter integrity and differentiate occurrence of cumplications with rclationJo patterns of u x . Number of ptInCNRS was &fined as the t d number of times the portal xptum and overlying skin was penetrated with a Huber needle. This was uscd to asxss vptum integrity and cutaneous reactions and to diffemntiav the occurrence of complications with relation to paneras of use. All pafienu who had PAC system implanted continued to be monitored on sndy. even after having completed the prescribed course of chemotherapy.

RESULTS

Performance The PAC venous access system was implanted for a total of 7,240 days, in use for a total of 1,435 days, and was subjected to a total of 525 needle punctures. Septum integrity was excellent; there was no evidence of septum damage or leakage, and no necrosis of the overlying skin was noted. No inadvertent extravasation of heparin. contrast dye, or vesicant chemotherapy occurred. and no significant movement of the portal from the original site of implantation was observed. Resistance to fluid flow was minimal with the venous PAC. and infusion rates of up to 300 mUhr were achievable with a number of commercially available infusion devices. Clinical data for the 32 patients who received implants with each of the two venous configurations are outlined in Table 2 and separated according to principal pattern of use. Because of the open-label design of the study, patient-related variables such as diagnosis, age, extent of disease, or chemotherapy treatment could not be controlled and thus affected measured parameters for system performance. In addition. the LB configuration was not available before May 1983, resulting in a relatively shorter median implantation time for LB patients as opposed to SB patients. There were more patients in the SB group with favorable diagnoses, and this resulted in more long-term survivors in the SB group than in the LB group. It is apparent that the majority of longterm survivors were receiving short-term chemo-

therapy. resulting in a significantly longer rnedian implantation time (552 days) in the short-term subset as compared with the other three patient groups. Despite the shorter median implantation time in LB patients, median times of use were comparable in both groups. This could be explained by a greater proportion of the LB patients with long-term patterns of use. Only seven of 16 SB patients (43.89) received TPN or continuous chemotherapy. in contrast to ten of 16 LB patients (62.5%) receiving similar therapies ( P = .07). The median implantation time, and the time of actual use were both longer in LB patients than in SB patients in this long-term subgroup. This could be related to the somewhat higher incidence of early death and the significantly higher incidence of complications in SB patients in contrast to LB patients. Complicariom Mild local e q d w n a and swelling of the skin overlying the portal was frequently noted postop eratively, but resolved in dl cases within 72 hours of implantation. Mild to moderate surgical discomfon was also noted and lasted from 2 to 7 days, but in no instance interfered with the use of the device. S e p m damage or leakage was not observed during the course of this study; however, local complications included irritation of the skin overlying the PAC in two patients receiving TPN via an indwelling Huber needle. Both patients recovered following local measures and Huber needle replacement. Dislodgement of an indwelling needle occurred in one patient and resulted in extravasation into the subcutaneous portal pocket. This patient was receiving IV fluids and recovered uneventfully. Resistance to fluid flow was observed in I I patients implanted with the venous PAC system after multiple uses. In four cases, a repeat flushing procedure was uneventful and implied that incomplete needle insertion was the cause of flow resistance. In seven patients (21.5%), resistance to flow was persistent, and the administration of radiographic contrast dye through the PAC confirmed catheter occlusion as the cause of flow resistance in all cases. Five patients underwent catheter declotting with urokinase (Abbon Laboratories. North Chicago, Ill) via a procedure previously de~cribed.~ with patent flow reestablished in two cases. Four patients under-

STRUM R AL tton. When u d f a mernlgh IOlong-term penods of lime. the bent Huber d k m a s tmcncd Into the porul and firmly affixed to the overlying & ~ nrc wdl be devnbed

Dressing an Indwelling Huber Needle The length of rhc bent Hubcr needle to bc used was individually determined f a each paticnr by the thickms of the skin and subcutaneous tissuc overlying the p n a l . In most cases. the 25.4 mm-bent needle uas of wffuicnt length to pcnevate fully the septum and overlying s k ~ nmd mult in close proximity of the needle junction to the P i n aufwe. The overljing skin in some o h patients was rhK1; emugh to require the uu of the 38.1 mm-bent needk. In dl clscr. thrt v i m of the ncsdk projecting above the skin mfxre -as m i n i m a by the usc of the appropriate needk Length and. if requid. rcmfornmenr with ow or more sterik p a w pds. To provide an added barrier against infection. the nedk c m y site was coated uith a thin layer of antibiotic oinmnc. and the skin ovalying the paul was protected with a Ucrik m m p c n t dressing. The h i n g was firmly placed ovcr tht Hubcr m d l e and its junction wirh the connection tubing tod m n f a u d with hjpdkrgenic adhesive tape. This was don lo prevm inadvertent needk dislodgement. or migration from the porul by socalled plastic cnwl. which could result in s e p m damage a drug cxtnvtwtion.

Carheter Mainlenance I

through fascia

Fig 2. (A) For centml v e w access, the PAC portal war surgically implanted into a subcutaneous pocket created in the right or left infmclavicular space. The catheter wcn i n d u c e d into the superior vena cava with the aid of a guide wire via the external jugular or cephalic vein. The catheter war cut to the proper length, tunneled subcutaneously to the pocket, and connected to the portal with the locking ring. (6) The porial was s e c u d in position to underiying fascia with four 2.0 Ethibond sutures on an 0 5 4 n d i e at the four suture loop rites.

by triangulation. However. in mmc obese patients. the paul was more difficult to locate. In such uses. India ink was used to tanw the skin overlying the paul. d m b y outlining the p o d perimeter. Huber needles were specirwd f a injection of tht PAC to minirnizt polmtid septum durugc. a coring. The feel of the needk &ing through the ~ - p o c uvptuh l was sufficiidy characteristic ro preclude indverteM injection of drug into the subcutaneous tissue overlying the PAC. Special carc was taken during the injection rsocuhm to nuintin fluid flow into the P A C . - c&stanl ~ k i t i v e prcuurt was applied on the syringe plunger w k n withdrawing the m d k from the p n a l to prevent possible aspiration of blood into tht system. For simpk heparin flushing a bolus injections of c h e m therapy. a direst approach was used in accessing the PAC. using r syringe and straight Hubcr ncedk. For patients meiving TPN or intermittent or long-term ckmabcrapy infusions. r Hubcr needk bent 90 d e g r m was used in conjunction with a sbon length of connection tubing to facilitate drug or TPN adminima-

Catheter parency m s rnainuined in accordance with sndy protocol through a routioc fllnhig scheduk md snict adhcrcnce to prescribed ca r c porrdurc.' If the system was noc used for drug or TPN d m i n i s n t i i within 24 lo 48 houri afkr placement. the PAC f l u with 5 mL h e p u i n i d saline of 200 UlmL. The PAC was using an empiric c a r t n ~ i a n flushed with 5 mL Dornul d i n e bcforc each use to mify the patency of the systrm and IO c k u residual hcpann to avoid potential d r u g - d ~ gi o t ~ i o a sIf. used f a inttrminent c h e w thenpy, the PAC was flusbed n t h heparinid d i n e aftaeach use and once every 30 dap d a u m c n t was adminiwend kss frequently. If used f a TPN a long-krm chemochmpy. the PAC was flushed arr u c h week the t i m the indutlling dressing wcrc Hubcr medk, conrrrtioa ruling. md &k changed.

Patterns of Usc CadKltr applicuiom w r e mzd uniformly distributed within

the two subgrwps; the n u mof patients in h e LB p u p had thtir PAC plrccd f a prrpsa d TPN. vhich rtW O I Q higher propmion d p a k h~m a~g grsbointcstirul obsuctioo resulting han rWorninrl m d ' a pelvic curiaanuasis. H m e v a.all prtienls rectivd marrnc with cithcr interminmt a mrinuous infusion &mahapy. ToassessIhc~arnancxmdmpliuIioontewithdK PAC under Ihe va.rias c o n d i h descrikd. IWO brod crileria were used to definc prrcrm d use. Patients who m i v c d their therapy as m i n t n m h a x injection or infusioa were c l u i f K d into the shori-term a b q a y . In these paticnu, dw PAC wls wc in use for I - to 4-veeL pcriodr. with the frtqvncy of uu dependent on the type d c l u m b m p y regimen used. In dl cases. the system w s hshcd with heparinid d i n a 3O-day use. whvhever occumd fust. Rtienls intervals or afiaCAI who wcrt receiving TPN a caacinuous infusiom of c h e m o d m -

went surgery, with thrombectomy and replacement of the clotted catheter. One patient was moribund at the time of venous catheter occlusion and was not considered to be a surgical candidate. Examination of the explanted catheter in all cases revealed the presence of organized thrombi occluding the catheter tip at the distal end. All seven occlusions were noted to occur in patients who received the SB venous catheter implant (seven of' 16, 43.8%). which occurred a median of 64 days post-implantation and after a median of 34 days of use. Five of seven occlusions (71.4%) occurred in patients receiving TPN or continuous chemotherapy for a median of 54 days of use (Table 3). In contrast, no occlusions occurred in patients who received the LB catheter implant (zero of 16,096. P < .01). This Table 2.

occurred after a median implantation time of I I2 days despite the significantly greater frequency of TPN use in the LB subgroup. Severe neutropenia following chemotherapy for small-cell lung carcinoma developed in one patient with an LB catheter. Clinical sepsis developed, with peripheral blood cultures positive on two occasions for Klebsiella pneumoniae and Candida albicans. Despite treatment with broadspectmm antibiotics and antifungal therapy, the patient died on the tenth postoperative day. Antemortem studies of the explanted catheter failed to reveal fungal elements, and catheter tip cultures were negative. DISCUSSION

The results of this study indicate that the implanted PAC system is able to provide reliable

Patient Data With the Venous PAC System LB System

SB System

No. of Time *I( Patient No. Implant U d

.

PAC

Corn-

,,+-plicoriom

tion

Potimt Status

NO.

No. of PAC Cornpl.co-

,+

l - 8 ~

Implant

U d

tion

Status

5 19 8

no no no

deod

7

no no no

olive

ticas

20 10

dead dead dead

~ r i

10 11 13 14 Total

Median

Total

69 9

1 36 22 3 9 5 25 15 12 10 138

no no

deod dead deod deod dead &ad

no

alin

no no

olk

no

0ri

no

na no no'

dead

Median NOTE: Potienh 1 to 16 received 58 cathcta impkds; potienh 17 to 32 mtircd LB catheter +h. cocrglicoth n o d . or potiart &&eviatiom: Pt, patient; implant time, d a p PAC i m p l o r d t)rargh Feb 1. 1%. death; time d, PAC octually used; no. of PAC injections, number of skidseptum p m d ~( ~ ),; day rcpkccd corhctu u d .

'Septic death, 10 clays postimplantation.

P A C A N IMPROVED METHOD FOR VENOUS ACCESS table 3.

Thrombotic Cornplicationr in Patients With SB Systems

No. of Patient No.

Primary Use

lmplont

Used

tions

704 64

275

73 9

CT

CT TPN, CT TPN, CT TPN, CT TPN, CT TPN, Cl T~tol

Medi

34

NOTE: Catkter thrombosis war the maior complication limiting clinical use of the PAC, and all occur~edin p t k t r who received S8 r y r t m implants. Five of wen thrombotic coqdicotions occw~edin potientr receiving TPN a d o r continuous chemotherapy. Surgical m e c t i o n consisted of thrornbectomy ad surgical replacement of catheter. Abbreviations: implant time, number of d q s PAC implanted until occlusion occurred; time used, number of days PAC octwlly used until occlusion occured; no. of PAC injections, nmber of skidseptum pKhrres until occlusion occvrred; CT, chemotherapy. 'Uroki~sefailure. tPotimt rnoribvnd ond surgg. not irdicoted. SUrdrinou success (5,000 IU instilled into PAC for one or two ottempts). 5Urokimu use not &tempted.

venous access to cancer patients in whom all peripheral sites have been exhausted. Our clinical experience'with the PAC appears similar to those of others who investigated other implanted devices'' and to those of Bothe et a14 who also evaluated the PAC system. However, in contrast to previous studies, we have reported data regarding long-term performance and complication rates, including five patients who have had a PAC system implanted for more than 2 years. Ten study patients (five with SB catheters and five with LB catheters) are alive with functional implanted systems after a median implantation time of 623 days (range, 165 to 729 days). A number of approaches have been tried in the past to provide intermittent to long-term vascular access. These include arteriovenous fistulas.' bovine heterografts.' and external catheterization of the right atrium via the peripheral or subclavian routes with various types of silicone or Silastic catheter^.^'^ Of these, the Broviac" and Hickman" catheters are perhaps the most commonly used. Although efficient, they possess problems with infection and occlusions that generally limit their use to approximately 100 days.".'' They also require a tedious maintenance schedule, and because of the externally exposed injection site,

inhibit activities such as swimming and bathing besides being cosmetically unamctive. Patient acceptance of the PAC system has been excellent. The cosmetic appearance of this implanted access system has been favorably received, especially by female patients. Of greater importance, however, is that the PAC system has eliminated the tedious search for suitable peripheral veins. In addition. the 1.02-mmdiameter of the LB system currently in use also allows for blood sampling. Pain, anxiety, and apprehension previously experienced by our patients with respect to blood sampling and chemotherapy administration have virtually been eliminated. Complications with the PAC system were limited primarily to patients receiving TPN with the SB catheter configuration. Since May 1983, this system was replaced with the LB catheter, which appears to have eliminated thrombotic complications with the PAC. However, the SB systems appear to perform well for short-term intermittent therapies. Thrombosis developed in only two of nine patients (22.2%)after a mean of 462 days of implantation and a mean of 20 days of use. In two of five cases in which thrombolytic therapy was used, 5,000 to 10.000 U of urokinase instilled into the PAC effectively declotted

,

the system, thus avoiding the need for surgical intervention. The performance of a similar implanted catheter system was tested in 24 patients by Gyves et In this study. catheter occlusion was noted in five of six patients (83.3%) who received a 0.38-mm-diameter catheter implant after a median of 180 days of implantation. However, no thrombotic complications were noted in 18 patients with a 0.63-mm internal diameter catheter implant after 300 days of observation. This is in agreement with our results with the PAC system and suggests that risk for catheter occlusion is related to the bore size of the catheter used. The potential for drug extravasation with vascular access ports has been documented in the recent literat~re.'~ In this study, five of ten patients who received implants with various implanted catheter systems suffered drug extravasation into the skin overlying the portal during overnight infusions. In three cases, resultant tissue injury was severe enough to warrant surgical removal of the access port. However, in our experience with the PAC system, risks for extravasation with this device appear to be much lower. occurring in one of 32 patients (3.1%). This is similar to results of another study of the PAC device in which three of 74 patients (4.1%) experienced drug extravasation.' We attribute our lower rate of drug exmvasation with the PAC to strict adherence to our catheter care protocol' and weekly changes of the indwelling Huber needle in patients receiving TPN or continuous chemotherapy. Another important factor relates to the specific design features of the PAC system that differentiate it from other vascular access pons. The silicone septum and portal chamber of the PAC device require significantly deeper needle penetration, thereby providing greater friction between the needle and septum. This feature makes inadvenent dislodgement of an indwelling Huber needle less likely to occur. thus reducing risks for extravasation. Controlled trials are required to determine the relative risks of drug extravasation among other commercially available vascular access ports, in conjunction with rigid adherence to a catheter care protocol. Infectious complications with the PAC in our study were negligible and confirm initial reports of a low incidence of infection with such implanted systems. Strict aseptic technique in all

aspects of care of the PAC as well as uniform surgical technique by a single surgeon (A.K.) may account for our favorable infection rate. However, patient selection in our study population would also appear important since other factors such as underlying disease status and immune suppression are known to affect the potential for infection developing in a cancer patient. In earlier studies in which Hickman catheters were used in granulocytopenic patients, the risks of infection and sepsis developing were obviously much higher.I6 Since only one patient in our study population became severely neutropenic during chemotherapy, this bias may have favorably influenced the incidence of infection in our study. Further studies of implanted catheter systems involving a larger patient sample with varying clinical circumstances are required to determine the true incidence of infectious complications with totally implanted catheter systems. The availability of the LB catheter configuration allows the PAC system to be used for blood sampling. as well as the administration of blood or blood products. an advantage that was not present with the earlier SB catheter configuration. However, we did not use the LB catheter extensively f a phlebotomy purposes except in patients who had no other peripheral sites. It should be noted that strict adherence to protocol was required for optimal use of the LB catheter for phlebotomy purposes. If excessive suction was applied, the thin-walled design of the catheter was noted to collapse and result in the inability to withdraw b l d from the PAC. even though the lumen remained patent for drug administration or TPN. This phenomenon, referred to as a partial occlusion, has also been observed to occur in patients with Hickman catheters." We conclude that the PAC system is a safe and reliable method of venous access for intermittent or extended periods of time and appears to be a reasonable alternative to external catheter placement in patients receiving cancer chemotherapy or TPN. Complications consisted primarily of a high rate of system occlusion, which appeared to be related to the use of the SB catheter system. For patients receiving chemotherapy or TPN, the complication rate with the LB catheter configuration presently available would appear much lower. Comparative trials involving a large sarnpie size are required to determine relative per-

PAC-AN

IMPROVED METHOD K)R VENOUS ACCESS

formance, complication rates, and patient preferences between totally implanted access ports and external venous catheters. REFERENCES 1. Coares A, Abraham S. Kayc SB. a d: On Lhe receiving c n b P a t i e n t perception of the s& effects of cancer chemotherapy. Eur J Cancer Clin Oncd 19:20>3X. 1983 2. Gyves J. Ensminpcr W. Nidchuber J. cf d: Totally implanted system for inmvenous chernochmpS in patients u-ith m r . Am J Med 73:841-845. 1982 3. Ecoff L. Baronc RM. Simonr. RM: Implanubk infusion pm (Pofl-A-Cath). Journal of the N a t i d lnmvenous Therapy Association 2:4&408. 1983 4. h h e A. R c c i m W. A m h i n o JJ. a 11: lmplanrable central v e m s access system. Am J Surg l47:MS-569. 1984 5. McDcmwd JE. Tracy IT. S u m SB: P a r - A C a t h policies and procedures. A manual for physicians. pharmacists. and n w u s . Piscatrway. NJ. Pharmacia Labartoriu;. 1985 6. Lawson M. Bottino JC. Humbisc MR. a A: The usc of urdrinase to mslorc patcncy of a c l u d e d c t n t n l venous carha e n . Am J lntra Ther Clin Nuu 9:-32. 1982 7. Wade JC. Newman KA. Schimpfi SC.cf al: Two methods fa improved venous access in acute leukemia palienu. JAMA 246:140-144. 1981

8. Lempert N. MacDowell RT. Karmody A. el al: Vascular rrrss f a cancer chemotherapy. Cancer 43:1934-1936. 1979 9. h i n o J. McCrcadie KB. Groschel DH, el d : Long-term intnvenus lhcnpy with pcriphcrally inserted silicone elastomer central venous calhetcrs in patients with malignant disuus. Cancer 43: 1937-1933. 1979 t vascular access: The 10. Lzird JJ: R n n u ~ n subcutaneous cJumo shunt. C m J Surg 25:690-692. 1982 11. Brovir IW. Cole JJ. Scribncr BH: A s i l i m rubber mill carhrter forprolonped p u c n w a l alimentation. Surg Gynecd Obsla 136:602-606. 1973 I?. H r k m RO. Buckncr CD. Clift RA. el d: A modified right ahid cllheter for access to Ihe venous system in marrow tnmpiun recipients. Surg Gynecol Obrtet 148:871-875. 1979 13. Tbomrc JH. M v A n u r RI. Rcrcc GE, et d: HickmanBrwiY calhc~ertindicationsand rcsullr. Am J Surg 140: 791-796. 1980 14. &gala JE. Maher K. Cherry ID: Risk of infection assxiued airh the use of Broviac and Hickmancatheten. Am J Infect C d 1&17-23. 1982 IS. Reed WP. Newman KA. Applcfeld MM. ct 11: Drug eamvasuion as a complication of venous access pons. Ann hem Med 102:788-789. 1985 16. R e d WP. Ncwman KA. De Jongh C. el A: Prolonged vcmous mss for chemochenpy by means of the Hickman catham. Cancer 52:18>192. 1983