中華放射醫誌 Chin J Radiol 2004; 29: 61-66 61

The purpose of this study is to evaluate peripher-ally inserted central catheter (PICC) as a solevenous access device for management of patientswith newly diagnosed leukemia.

Forty-four PICCs (4-French, single lumen) wereplaced in 41 newly diagnosed acute leukemiapatients mostly complicated with active systemicinfection and bleeding tendency. Sonography-guidedvenipuncture using a 14-gauge needle sheath wasperformed because of no grossly visible forearmperipheral veins. PICC insertion sites were basilicvein in 14, brachial vein in 12, cephalic vein in 13,antecubital vein in 4, and in right internal jugularvein in one patient after failed antecubital approach.Catheter tips were placed at right atrium and supe-rior vena cava junction in all. The catheters wereused for infusion of fluid, chemotherapeutic agents,blood products, and total parenteral nutrition.

No hematoma was encountered. The PICCs wereused for a total 3219 catheter-days, with a meanindwelling time of 73.2 days (range 1-247 days).Complications occurred in 18 catheters (18 of 44,40.9%): puncture site infection in 4, phlebitis in 2,sepsis in 3, catheter clotting in 4, central venousthrombosis in 2, cannulated vein thrombosis in 1,catheter kinking and hub leakage in 1, and acci-dental removal by nursing personnel in 1. Inductionchemotherapy was accomplished in 75% of PICCs

(33 of 44), with an average of 1.9 courses (range 1~5courses).

PICC is safe and useful as an interim centralvenous access device for management of critically ill,newly diagnosed adult acute leukemia patients com-plicated with active infection and bleeding tendency.

Key words: Acute leukemia; Peripherally insertedcentral venous catheter; PICC; venous access

Patients with newly diagnosed acute leukemia tra-ditionally require large bore central venous access forfrequent fluid infusion, blood transfusion, blood with-drawal for laboratory tests and parenteral nutrition [1].To accomplish the above purposes, surgicallyimplanted chest ports or Hickman catheters are oftenused [1,2]. Unfortunately, chest ports or Hickmancatheter insertion in newly diagnosed leukemiapatients are associated with high infection rate andaccess site hematoma formation due to impaired whiteblood cells function and severe thrombocytopenia[2,3]. A prior study on a general patient populationindicates that the small caliber (4-5 French) peripher-ally inserted central catheter (PICC) is a useful andcost-effective alternative venous access device to con-ventional central venous catheter [4]. The small caliberperipherally inserted central catheters (PICCs) are the-oretically well suited for use in critically ill acuteleukemia patients because their insertion through theperipheral vein eliminates the complication of pneu-mothorax and avoids surgical dissection which aremandatory during Hickman catheter or chest portinsertion that may result in access site hematomaformation [5,6,7]. However, little data was collected inthe use of small caliber (4-5 French) PICC as a venousaccess device for management of critically ill acuteleukemia patients [4,8,9,10]. The purpose of this study

Reprint requests to: Dr. Kee-Min YeowDepartment of Diagnostic Radiology, Chang GungMemorial Hospital.No. 404, 11th Floor, Chang Gung Medical CommunityVillage, Kwei Shan, Tao Yuan 333, Taiwan, R.O.C.

The Use of Peripherally Inserted CentralCatheter (PICC) in Adult with AcuteLeukemiaSUNG-YU CHU

1 CHEN-HSIEN WU1 JIAN-LUN WU

2 CHEN-HONG TOH1 KUANG-TSE PAN

1

JENG-HWEI TSENG1 CHIAN-FU HUNG

1 KEE-MIN YEOW1

Department of Diagnostic Radiology1, Chang Gung Memorial Hospital, Chang Gung UniversityDepartment of Rehabilitation2, Mackay Memorial Hospital

is to investigate the feasibility of using PICC as analternative venous access device to Hickman catheteror chest port for management of patients with newlydiagnosed acute leukemia.

MATERIALS AND METHODS

From June 1998 to Sep 2000, 41 patients, 20males and 21 females, aged 42.5-53.3 years (mean47.6) with newly diagnosed acute leukemia werereferred from a team of hematologists for insertion ofsmall caliber PICC due to difficult peripheral venousaccess. The mean platelet count of these patientsbefore PICC insertion was 4.8 × 104/µL (range 7000-282000/µL, CI 95%, 3.35-6.27). Each of the 41 con-secutive patients received one catheter on initialinsertion. Thirty-eight patients had one catheter, andthree received two after their initial catheters wereclotted or because of insertion sites complicationsoccurred. The PICCs were used for infusion ofchemotherapeutic agents, fluid infusion, blood transfu-sion, blood withdrawal for laboratory tests, and par-enteral nutrition during induction of acute leukemiaand maintenance of disease during remission.

The catheters were inserted using a standardsonography-guided technique because there were novisible peripheral veins in the forearm regions of thesepatients. A tourniquet was tied around the arm duringsonographic scanning of left forearm veins at 1-2 cmbelow the antecubital fossa using a 7.5 MHz high-res-olution sonographic probe (Acuson, Mountain View,CA, USA). A left antecubital vein (either the brachialvein, basilic vein or cephalic vein) was selected forvenipuncture. Left forearm insertion was routinelyselected for fear of inconvenience of right armmovement because all our patients are right handed. A20 cm2 wide area around the planned insertion site wascleansed by povidone iodine solution and draped.After local anesthesia, venipuncture was performedusing a 14-gauge needle sheath under direct sono-graphic guidance [9]. A 4-French single lumenGroshong tip PICC (Bard Access System, Salt LakeCity, Utah) was inserted into the vein via a plasticsheath after removing the needle stylet. The PICCswere inserted blindly up to the 50-cm mark indicatedon the catheter. The final position of catheter tips waschecked using fluoroscopy (Fig. 1,2). The introducersheath was then removed. The external portion of

Use of PICCs in acute leukemia patients62

Figure 2. Chest radiograph confirms the location of thePICC tip (arrowhead) at the junction of the right atriumand superior vena cava.

Figure 1. A PICC enters the basil ic vein at leftantecubital fossa below the elbow (arrow). Its tip(arrowhead) is positioned at the superior vena cava level.

PICC was trimmed to appropriate length andconnected to catheter hub (Fig. 3). Only the 4-Frenchsingle lumen PICC with a hemostatic Groshong slitvalve (Fig. 4) (Bard, Access System, Salt Lake City,Utah, USA) was used in this study. After successful

insertion, the PICC function was tested by smoothblood withdrawal and saline infusion. The catheterentry site was covered by sterile gauze and thenaffixed to patient’s skin using a 10 × 12-cmTegaderm® (3M Health care, St. Paul, MN, USA). Noprophylactic antibiotic was given before or after theprocedure and no antibiotic ointment was applied tothe antecubital venous puncture sites. The gauzecovering the catheter entry site was changed daily bynursing personnel.

In this retrospectively study, the medical chartswere reviewed. The blood counts before theprocedure, date of catheter insertion, insertion site,date of removal, reasons for catheter removal,chemotherapy received, and complications werereviewed. The catheters were followed until they wereremoved or when patients expired.

DefinitionA “success” at initial placement of PICC was

defined as proper insertion of a PICC with its tipplaced at junction of right atrium and superior venacava. Sepsis is defined as bacteremia of unknownsource or developed soon after PICC insertion withpositive blood cultures. Cellulitis over the insertion-site of PICC was considered a “local infection” ifthere is no sign of systemic infection. A venous throm-bosis is diagnosed when a thrombosed and distendedvein is palpable but not tender. Phlebitis is defined as

Use of PICCs in acute leukemia patients 63

Table 1. Subtypes of patients with acute leukemia

Subtype Number

AML* M1 M2 M3 M4 M5 M6 M7 351 6 14 9 3 1 1

ALL** L1 L2 L3 54 0 1

Mixed AML and T cell-ALL 1Total 41

* AML: acute myeloid leukemia** ALL: acute lymphoblastic leukemia

Table 2. Catheter-related complications

Classification Number

Sepsis 2 (4.5%)Suspicious sepsis 1 (2.3%)Puncture site infection 4 (9.1%)Central venous thrombosis 2 (4.5%)Cannulated vein thrombosis 1 (2.3%)Phlebitis 2 (4.5%)Catheter occlusion 4 (9.1%)Catheter leakage and kinking 1 (2.3%)Accidental removal 1 (2.3%)

Total 18 (40.9%)

Figure 3. This is the forearm picture of a 45 year-old manwith acute myeloid leukemia. A 4-French PICC enters thecephalic vein just below the elbow. The patient has aplatelet count of 15000/µL and spontaneous forearmecchymosis are noted. No puncture site hematoma isnoted after a successful sonography-guided venipuncture.

Figure 4. The diaphragm depicts hemostatic function ofthe Groshong slit valve at the closed distal tip of a PICC.Top, the valves open toward the venous lumen duringinfusion of fluid through the catheter. Middle, the valvesare sucked into catheter lumen during blood withdrawal.Bottom, the slit valve closes when the catheter is not inuse preventing blood entry which can cause catheterclotting.

Infusion-Positive Pressure

Aspiration-Negative Pressure

Close

painful sensation and erythema along the insertedvein. Blood clots within the catheter or twisting of thePICC catheter is defined as catheter occlusion.

RESULTS

Forty-three PICCs were inserted in 41 leukemiapatients with a 97.7% (43 of 44) technical success rate(Table 1) and one catheter was inserted via rightinternal jugular vein after a failed antecubital fossainsertion. The insertion sites were basilic vein in 14,brachial vein in 12, cephalic vein in 13, antecubitalvein in 4, and right internal jugular vein in one. Themean catheter indwelling time was 73.2 days (range1~247) and for a total 3219 catheter days. Thirty-threePICCs (75%) have been used for inductionchemotherapy for an average of 1.9 courses (range1~5 courses).

No hematoma was noted at puncture site, andminor oozing was controlled by local compression orstopped after subsequent transfusion of bloodproducts. No prolonged bleeding was noted.Complications directly related to PICC insertion occurin 18 catheters (18 of 44, 40.9%), translating into 5.59per 1000 catheter days. The average duration fromPICC insertion to development of complication was58.8days (range 3-225). There were 7 infectious com-plications (7 of 45, 15.6%) (Table 2). Four local infec-tions (4 of 44, 9.1 %) occurred and the mean durationof onset was 14.8 days (range 4-36). Three systemicinfections (6.8%) were noted and the mean duration ofonset was 103.3 days (range: 52-183). In the patientswith systemic infection, one was confirmed bycatheter tip culture which grew Staphylococcusaureus. Another patient with suspected systemicinfection was unable to be confirmed by blood culture.The other patient with systemic infection had con-comitant infection of catheter insertion site, and hisblood culture grew Enterococcus coli and Klebsiellapneumoniae. No wound culture was done in the lastpatient because there was no pus or discharge. Theoverall infection rate was 15.9%, translating into 2.17per 1000 catheter days.

One PICC (1 of 44, 2.3%) had a kink at theexterior segment of the catheter and hub leakagewhich developed on the 225th day after PICCinsertion. Four PICCs (4 of 44, 9.1%) were occludedby blood clots shortly after blood withdrawal forblood tests or due to spontaneous reflux of blood intothe catheter via the incompetent Groshong valves.Catheter occlusions developed on an average of 36.5days after insertion (range 22-88 days). Two of the 4occluded catheters were recannalized by urokinase

administration. Rapid development of central venousthrombosis occurred in 2 PICCs (2 of 44, 4.5%) andoccurred after a mean duration of 4.5 days in use(range 3-5). Thrombosis of the cannulated veindeveloped in a patient on the 172th day after initialplacement. Two patients had catheter induced phlebitison the 19th and 24th days after their insertion. Onecatheter became dislodged on the 162th day due toaccident removal by nursing personnel.

Fourteen of our patients (14 of 41, 34%) died ofacute leukemia during the study period.

DISCUSSION

Lately, there is a renewed interest in PICCbecause it is five-times less costly than conventionalcentral venous catheter (CVC) [5]. Smith et al [5] intheir review on 838 venous catheters (283 CVC, 555PICC), observed no increase in infectious complica-tion in PICC as comparison with in conventionallarger central venous catheter. Alhimyary et al [11] intheir study on parenteral nutrition via 135 PICC linesfor 126 patients shows no significant difference incomplication rate in comparison to central venouscatheters. In a high risk group patients with acuteleukemia, our PICC infection rate is 2.17 per 1000catheter days in contrast to 2.14 per 1000 catheterdays in a general medical group and 2.25 per 1000catheter days for the HIV group obtain by Ng et al [4].Bakker et al [7] report a higher infection rate at 4.78per 1000 catheter days for Hickman catheter inpatients with hematologic disorder.

In acute leukemia patient, easy bleeding and highinfection rate are the main concerns. Ng et al [4]report that coagulopathy is not a contraindication forPICC insertion even 14.6% of their patients requirevenous cut down to enable catheter insertion. In theiropinion, a pre-procedural coagulation study may notbe required in leukemia patients because blood testsand blood transfusions are frequently performed inacute leukemia patients. In other words, bloodproducts may be transfused after PICC insertion if thewound oozing is difficult to stop by manual compres-sion. However, we try to minimize the risk of bleedingby pre-procedural transfusion of fresh frozen plasmawhen platelet count is below 20000/µL while thewound is routinely compressed for a longer time.Loughran et al [6] obtain a phlebitis rate of 9.7%. Intheir study, the early phlebitis occurs on an average of3.7 days after insertion. The late phlebitis occurs on anaverage of 18.27 days after insertion. Only 2 latephlebitis occur in our series resulting in an incidencerate of 4.5% and no early phlebitis is noted in our

Use of PICCs in acute leukemia patients64

study. It is reported that an experienced operator isimportant in minimizing the incidence of complica-tions and to shorten operation time [4]. A gentle andsmooth insertion of PICC may reduce the incidence ofmechanical phlebitis which is thought to be related toexcessive movement of catheter during insertion [5]and manipulation or movement at excision site [12]. Inaddition, a proper patient education on daily PICCcare should be important. In our hospital, the radiolo-gist plays a key role in the insertion of PICC. Sincethe nurses are not well-trained in the PICC daily care,the occlusion rate of PICC is still high. We think thephlebitis and occlusion rate of PICC will decrease inthe future.

A similar study of PICC in acute myeloidleukemia patients was proposed by Strehilevitz et al[10]. The result of this study revealed a higherphlebitis rate (12 of 40, 30%) and sepsis rate (17 of40, 42.5%) than our result. We think usage of sonog-raphy-guided procedure in our study is the key role todiminish the complications, especially in the develop-ment of phlebitis.

In 1974, Ryan and colleagues [13] reportedinvariable occurrence of thrombophlebitis after ante-cubital-vein catheterization, which was thought to bedue to small caliber of vein and excessive movementof PICC at insertion sites. Jay et al [14] however,report a 3.7 to 4.5% incidence of venous thrombosis.In our study, the central venous thrombosis rate is4.4%. In Anthony’s series, a 23.3% of PICC relatedperipheral venous thrombosis is reported [15]. Ourvenous thrombosis rate is 6.8%, in which, two occur incentral veins and one in the cannulated peripheralvein. However, our venous thrombosis rate may beunderestimated because only symptomatic patients arefurther evaluated with sonography while the asympto-matic venous thromboses may be missed.

In conclusion, small caliber central venous accesscatheters inserted via peripheral forearm veins are safeand efficacious for intravenous therapy of newlydiagnosed acute leukemia patients. Our study showsthat small caliber PICC may be used as an initialvenous access device for the management of newlydiagnosed acute leukemia patients who requirefrequent blood drawing, blood product transfusion andparenteral nutrition. It has a comparable infectiouscomplication, venous thrombosis and catheterocclusion rates when compared to conventional largebore central venous catheter. However, proper trainingin PICC insertion technique and maintenance care aremandatory in order to obtain an acceptable complica-tion rate in these critically ill patients. ◆

REFERENCE

1. Krog MP, Ekbom A, Nystrom-Rosander C, et al:Central venous catheters in acute blood malignancies.Cancer 1987; 59: 1358-1361

2. Reilly JJ Jr, Steed DL, Ritter PS: Indwelling venousaccess catheters in patients with acute leukemia. Cancer1984; 53: 219-223

3. Rotstein C, Brock L, Roberts RS: The incidence of firstHickman catheter-related infection and predictors ofcatheter removal in cancer patients. Infection Control &Hospital Epidemiology 1995; 16: 451-458

4. Ng PK, Ault MJ, Ellrodt AG, et al: Peripherally insertedcentral catheters in general medicine. Mayo Clin Proc1997; 72: 225-233

5. Smith JR, Friedell ML, Cheatham ML: PeripherallyInserted Central Catheters Revisited. Am J Surg 1998;176: 208-211

6. Loughran SC, Borzatta M: Peripherally Inserted CentralCatheters: A Report of 2506 Catheter Days. Journal ofParenteral & Enteral Nutrition 1995; 19: 133-136

7. Bakker J, van Overhagen H. Wielenga J, et al:Infectious complications of radiologically insertedHickman catheters in patients with hematologic disor-der. Cardiovascular & Interventional Radiology 1998;21: 116-121

8. Eric KH, John B, Patricia S, et al: Prospective random-ized comparison of valved versus nonvalved peripheral-ly inserted central vein catheters. Am J Roentgenol1999; 173: 1393-1398

9. Chrisman HB, Omary RA, Nemcek AA, et al:Peripherally inserted central catheters: guidance withuse of US versus venography in 2650 patients. J VascIntervent Radiol 1999; 10: 473-475

10. Strahilevitz J. Lossos IS. Verstandig A, et al. Vascularaccess via peripherally inserted central venous catheters(PICCs): experience in 40 patients with acute myeloidleukemia at a single institute. Leuk Lymphoma, 2001;40: 365-371

11. Alhimyary A, Fernandez C, Picard M, et al: Safety andEfficacy of Total Parenteral Nutrition Delivered Via aPeripherally Inserted Central Venous Catheter. Nutritionin Clinical Practice 1996; 11: 199-203

12. Mazzola JR. Schott-Baer D. Addy L: Clinical factorsassociated with the development of phlebitis after inser-tion of a peripherally inserted central catheter. JIntraven Nurs 1999; 22: 36-34

13. Ryan JA, Abel RM, Abbott WM, et al: Catheter compli-cations in total parenteral nutrition. A Prospective studyof 200 consecutive patients. New Eng J Med 1974; 290:757-761

14. Jay RG, William CP: Venous Thrombosis Related toPeripherally Inserted Central Catheters. J Vasc InterventRadiol 2000; 11: 837-840

15. Anthony WA, Jocelyn LM, Daniel Brown, et al: VenousThrombosis Associated with the Placement ofPeripherally Inserted Central Catheters. J Vasc InterventRadiol 2000; 11: 1309-1314

Use of PICCs in acute leukemia patients 65

Use of PICCs in acute leukemia patients66

週邊植入式中央靜脈導管於成人急性白血病患者

朱崧毓1 吳政賢1 吳鑑倫2 杜振豐1 潘廣澤1 曾振輝1 洪建福1 饒啟明1

長庚紀念醫院　長庚大學　放射診斷科1

馬階紀念醫院　復健科2

急性白血病患者需要建立一條適當的靜脈輸液路徑，以往都是開刀植入大管徑（6-8Fr）的

Hickman catheter或是Port A catheter為主。我們嘗試著利用4Fr小管徑的週邊植入式中央靜

脈導管來達成此目的。

連續41位急性白血病的新患者（20位男性、21位女性，平均年齡為47.9歲）參與研究，

他們大多有全身性的感染及出血傾向（平均血小板為48000/uL）。我們由左手肘處植入了44條

中央靜脈導管在這些病人身上。由於在肉眼下看不到這些病人的手臂靜脈，導管的植入是在超

音波導引下進行，最後導管末端放至上腔靜脈處。這些導管將作為抽血、靜脈輸液、給藥及化

學治療。

放置週邊植入式中央靜脈導管的成功率為97.7%，置入導管的手肘部位並無任何的皮下血

腫發生。這44條導管放置在41位病人身上共3219天，平均每條導管放置73.2天。有18條導

管發生併發症（佔40.9%），這些併發症包括4條導管有穿刺傷口的感染、2條導管引起靜脈

炎、3條導管引起敗血症、4條導管發生阻塞、2條導管引起中央靜脈阻塞、發生放置導管的靜

脈阻塞有1條導管、1條導管的導管接頭發生滲漏及此導管發生扭轉的現象、1條導管在日常照

護時被意外拔除。這44條週邊植入式中央靜脈導管有33條（75%）曾經被作為化學治療的給

藥途徑，平均使用1.9次的化療療程。

週邊植入式中央靜脈導管併發症低於他人報告大管徑導管用於急性血癌患者的結果，而與

國外利用週邊植入式中央靜脈導管於一般病人的結果相近。所以對於此高危險群的患者不失為

Hickman catheter及Port A catheter的另一種選擇。

關鍵詞：急性白血病，週邊植入式中央靜脈導管，介入性步驟，靜脈