Short-term complications of central line placement in children with the human immunodeficiency virus

Short-Term Complications of Central Line Placement in Children With the Human Immunodeficiency Virus By E.R. Island, J.A. Church, and D.B. Shaul Los Angeles, California

Purpose: The aim of this study was to characterize the perioperative complications of central venous catheter placement in children infected with human immunodeficiency virus (HIV) Methods: A retrospective chart review was conducted of all central venous catheters placed by the surgical service into HIV-infected children from 1988 to 1998 at a large urban children’s hospital. Complications occurring within 1 month of catheter placement were analyzed for several host and environmental factors. Results: Forty HIV-positive patients underwent 60 central venous access procedures. Thirty-two of the patients were severely immunosuppressed. Eight catheter placements (13%) resulted in perioperative complications, including hemorrhage (n ⫽ 2), site infection (n ⫽ 2), catheter sepsis (n ⫽ 2), thrombotic occlusion (n ⫽ 1), and a pleural effusion secondary to catheter malposition (n ⫽ 1). Only 3 patients required catheter removal. There was no significant relation-

C

ENTRAL VENOUS CATHETERS have become commonplace in the treatment of children with malignancy, short gut syndrome, and any chronic illnesses that require frequent vascular access.1 These catheters provide easy vascular access for laboratory tests, central nutrition, and intravenous medication, while significantly decreasing the child’s anxiety associated with the pain associated with peripheral venous access. Central catheters are useful particularly in the human immunodeficiency virus (HIV)–positive population in which patients frequently require central access.2 In the pediatric HIV population, reducing the child’s anxiety surrounding phlebotomy is likely to result in a subsequent decrease in the risk of accidental needle sticks to the patient’s healthcare workers and family. Because of their compromised immune status, HIVpositive patients may be at unique risk for complications secondary to catheter placement. Previous reports in adults have documented similar risks of long-term infectious complications in AIDS patients when compared with oncology patients.2 No study to date has addressed the question of whether there is an increased risk of perioperative complications associated with central catheter placement in HIV-infected children. Therefore, this study was performed to determine whether these children are at unique risk for complications after central line placement.

ship between either hemophilia or thrombocytopenia and perioperative hemorrhage. No significant relationship was found between infectious complications and preoperative white blood cell count, absolute neutrophil count, CD4% and CD4#, suggesting that a patient’s compromised immune status should not be considered a contraindication to central venous catheter placement.

Conclusion: The complication rate of central venous catheter placement into HIV-infected children is low (⬍15%), but is still higher than that of the general pediatric population. With careful preoperative preparation this procedure can be performed safely, even in patients with advanced HIV disease. J Pediatr Surg 36:1777-1780. Copyright © 2001 by W.B. Saunders Company. INDEX WORDS: Human immunodeficiency virus, hemophilia, thrombocytopenia, CD4%, CD4#, absolute neutrophil count, white blood cell count.

MATERIALS AND METHODS Permission to perform the retrospective chart review was obtained from the Childrens Hospital Los Angeles institutional review board. The records reviewed included those of all HIV-positive patients who underwent central venous catheter placement over a 10-year period from 1988 to 1998. Some of the patients had more than one catheter placed over the 10-year study period. In these patients, each catheter placement was considered a separate event for the purpose of data analysis. All of the catheters were placed for central access by the surgical service in the operating room, the intensive care unit, or on the surgical wards. Peripheral infusion catheters (PIC lines) were not counted in the study sample. The records were reviewed for demographic data as well as clinical and immunologic staging criteria for HIV disease. At the time of line placement each patient was assigned a clinical and immunologic classification as defined by the Centers for Disease Control and Prevention (Table 1). The patients CD4%, CD4#, absolute neutrophil count (ANC), White blood cell count WBC, and serum albumin at the time of line placement were recorded. The records were reviewed for the primary indication for catheter, catheter type, site, placement technique (percutaneous or cutdown), and procedure location (operating room, ward, intensive care unit). Preoperative

From the Departments of Surgery and Pediatrics, Childrens Hospital Los Angeles, Keck School of Medicine, University of Southern California, Los Angeles, CA. Address reprint requests to D.B. Shaul, MD, Childrens Hospital Los Angeles, 4650 Sunset Blvd, MS #100, Los Angeles, CA 90027. Copyright © 2001 by W.B. Saunders Company 0022-3468/01/3612-0010$35.00/0 doi:10.1053/jpsu.2001.28821

Journal of Pediatric Surgery, Vol 36, No 12 (December), 2001: pp 1777-1780

1777

1778

ISLAND, CHURCH, AND SHAUL

Table 1. 1994 Revised Guidelines for the Performance of CD4ⴙ T-cell Determinations in Persons With HIV Infections Clinical Categories Immunologic Categories

N, No Signs or Symptoms

A, Mild Signs of Symptoms

B, Moderate Signs or Symptoms

C, Severe Signs or Symptoms

1. No evidence of suppression 2. Evidence of moderate suppression 3. Severe suppression

N1 N2 N3

A1 A2 A3

B1 B2 B3

C1 C2 C3

Data from reference no. 3.

factors, such as preoperative antibiotics and fever (temperature ⬎38.5°C) in the 24 hours before insertion were recorded and analyzed. Positive blood cultures in the week before the procedure were recorded. The records were reviewed for any complications such as infection, thrombosis, bleeding, occlusion, or rupture that occurred within 1 month of catheter placement. Data were entered into either a Microsoft Excel database (Microsoft Corporation, Redmond, WA) or GraphPad InStat statistics program (GraphPad Software, San Diego, CA), and statistical analysis then was performed. Statistical analysis consisted of the Student’s t test for continuous variables and Fisher’s Exact tests for categorical variables.

RESULTS

Forty HIV-positive patients, ages 10 months to 22 years, underwent 60 central venous access procedures during the study period. The patient population was 60% boys. The sample included 6 hemophiliacs (15%); the hemophiliacs were involved in 12 (20%) of the line placement procedures. With regard to the mode of transmission, 22 (55%) patients became infected with HIV through vertical (mother to baby) transmission, and 17 of the patients (42%) acquired HIV via blood transfusion. In one patient, the mode of HIV transmission was unknown. The state of HIV disease was quite advanced within our cohort. In our population, 77% of the patients met the requirements for class C HIV disease using the CDC clinical classification criteria, and 83% of patients had stage 3 HIV disease using the CDC immunologic classification criteria.3 Over the entire patient population, the mean ANC was 3,905 (range, 665 to 21,921), the mean CD4 count was 130 (range, 11 to 640; normal, 800 to 1,200), the average CD4% was 10% (range, 1% to 64%; normal, 32% to 56%), and the mean albumin was 3.2 (range, 1.7 to 4.5 g/dL). Four different types of catheters were utilized in this patient group (Table 2). The catheter placement sites

were the left subclavian vein (n ⫽ 27), the right subclavian vein (n ⫽ 11), internal jugular veins (n ⫽ 9), femoral veins (n ⫽ 5), and the external jugular veins (n ⫽ 5). The site of catheter placement was unknown in 3 of the procedures. Fifty-eight of the 60 procedures (96%) took place in the operating room. Catheter placement was accomplished using the Seldinger technique in 40 procedures (67%), and by venous cutdown in 18 procedures (30%). The technique used for catheter placement was not recorded for 2 of the procedures. Finally, the indications for catheter placement within this patient population included the requirement for long-term venous access (68%), long-term use of total parenteral nutrition (TPN, 25%), and the need for long-term use of intravenous medication (7%). Complications occurred in 8 of the 60 procedures. Table 3 lists the details of each complication. Only 3 of the complications resulted in immediate removal of the catheter. No relationship was found between patients with preoperative hemophilia or thrombocytopenia, and postoperative hemorrhage. Finally, there was no significant relationship between infectious complications and preoperative WBC, ANC, CD4%, and CD4# (Table 4). Patients who had central lines placed in the presence of a positive blood culture were highly likely to have a perioperative complication. Of 4 central lines placed in the setting of a positive blood culture, 3 were involved in a complication within 1 month of line placement (P ⫽ .006). However, these complications were not infectious in nature, therefore, the relationship between positive blood cultures and the complications may be coincidental. Preoperative fever and the use of preoperative antibiotics did not influence the postoperative complication rate.

Table 3. Complications Table 2. Catheter Type Type

No.

Complication

Number

Single lumen* Double lumen* Implantable port reservoir Percutaneous temporary Total

12 24 19 5 60

Bleeding Site infection Positive blood culture Occlusion Pleural effusion Total

2 2 2 1 1 8

*Cuffed tunneled SILASTIC姞 catheters (Dow Corning, Midland, MI).

Single Lumen

Double Lumen

1

1 1

1 1 3

Port

Temporary

1 1

2

1

1 2

CENTRAL LINE COMPLICATIONS IN HIV

1779

Table 4. Host Factors in Infectious Complications Entire Group

Number of lines Average WBC Average ANC Average CD4% Average CD4# Albumin

60 5,029 ⫾ 2,843 3,905 ⫾ 3,676 10% ⫾ 14% 130 ⫾ 256 3.2 ⫾ 0.6

Infections

4 2,231 ⫾ 374 3,603 ⫾ 1,034 2% ⫾ 2% 49 ⫾ 76 3.2 ⫾ 0.1

No Infections

56 4,016 ⫾ 2,903 5,113 ⫾ 3,766 10% ⫾ 14% 135 ⫾ 263 3.1 ⫾ 0.6

P Value*

.376 .066 .413 .192 .934

*Infection group versus No Infection group.

DISCUSSION

In this study, there was a 13% (n ⫽ 8) incidence of perioperative complications occurring within 1 month of central venous catheter placement. Although the overall incidence of complications is low, this number represents a substantial increase in complications when compared with the pediatric population as a whole. In 1998, Johnson et al4 reviewed 1,435 consecutive pediatric catheterizations over a 10-year period. These investigators reported a perioperative complication rate of 3.2%. The complications seen in their study did not appear to be directly related to the host’s immune status. Other studies looking specifically at line placement in the adult HIV population have reported slightly elevated but still low complication rates.3,5,6 However, by focusing on the immediate postoperative period, our study is the only one that addresses the question of perioperative risk. When looking at the specific complications encountered in this study, some interesting conclusions can be drawn. For instance, neither low platelet before the procedure, nor the presence of hemophilia, were associated with a perioperative hemorrhagic complication. This is most likely the result of appropriate preoperative preparation. The children were given appropriate platelets and coagulation factors before the start of their access procedure. Infection is always an important concern when dealing with an immunocompromised population. In our study, however, the incidence of perioperative infection was low. This may, in part, be caused by our focus on the immediate postoperative period. Infections that may have occurred 30 days or more after the procedure was performed were not recorded. Fernandez-Martin et al7 reported a mean onset of infection 58 days after central line placement in an adult HIV-infected population.7 They also noted that 7 (28%) of their infections occurred within 10 days of the procedure. Dega et al6 looked at 89 lines placed in 84 adult HIV-infected patients and found an infection rate of 28%. Five (20%) of these infections occurred within 5 days of catheter placement.6 Our study design captured 4 infections within 30 days of catheter placement. Two of the infections were in the form of positive blood cultures attributed to the central line, and 2 were cutaneous line site infections. All of these infec-

tions were treated, and none of these infections resulted in catheter removal. The patient’s WBC and T-cell status were not significant predictors of perioperative infection. This is consistent with previous published reports of line placement in the HIV population. Dega et al6 found no significant difference in mean lymphocyte count, CD4 count, and absolute neutrophil counts between patients who had subsequent infections and those who did not. In 1991, Gleason-Morgan et al8 reported that reduced CD4 T-cell counts were not predictive of catheter infection in a pediatric HIV population. It has been shown that a low ANC at the time of catheter placement predicts early catheter infection in pediatric oncology patients.1 The majority of patients in the current study had a normal ANC, and this may explain the low overall infection rate. Interestingly, the infection rate remained low in spite of the high prevalence of low CD4# and CD4% in this study population. This study found no significant relationship between catheter type, the number of lumens present, or the location of the catheter and subsequent perioperative infection. However, Dega et al6 reported that the rate of catheter infection in HIV-infected adults was related to catheter site and the type of catheter placed. Our findings are more consistent with those of Weiner et al9 whose multiinstitutional review of 1,141 line placements in pediatric cancer patients showed no significant relationship between either catheter type, number of lumens, or catheter location and subsequent catheter infection. The limitations of our study include a small patient population and our deliberately focused period of postoperative follow-up. However, some interesting conclusions about central line placement procedures in HIVinfected children can be drawn. All of these patients should undergo meticulous preoperative preparation. Particular attention should be given to optimizing the patient’s platelet count and coagulation factors before catheter placement. Also, the child’s history of recent infection should be researched, and the placement of long-term access devices should be avoided in the setting of a recent positive blood culture. When special attention is paid to the perioperative management, central venous access procedures can be performed safely in fragile pediatric HIV-positive patients.

1780

ISLAND, CHURCH, AND SHAUL

REFERENCES 1. Shaul DB, Scheer B, Rokhsar S, et al: Risk factors for early infection of central venous catheters in pediatric patients. J Amer Coll Surg 186:654-658, 1998 2. Henery K, Thurn JR: Experience with central venous catheters in patients with AIDS. N Engl J Med 320:1496, 1989 3. 1994 revised guidelines for the performance of CD4⫹ T-cell determinations in persons with human immunodeficiency virus (HIV) infections. Centers for Disease Control and Prevention. MMWR Morbid Mortal Wkly Rep 43:1-21, 1994 4. Johnson EM, Saltzman DA, Suh D, et al: Complications and risks of central venous catheter placement in children. Surgery 124: 911-916, 1998 5. Dimian CO, Bingham JS: The use of central venous catheters in patients with HIV. International Journal of STD & AIDS 6:455-456, 1995

6. Dega H, Eliaszewicz M, Gisselbrcht M, et al: Infections associated with totally implantable venous access devices (TIVAD) in human immunodeficiency virus infected patients. J Acq Immune Def Synd 13:146-154, 1996 7. Fernandez-Martin J, Salmon-Ceron, Leport C, et al: Septicemies sur catheters veineux centraux implant’s chez les malades atteints de SIDA. Med Mal Infect 22:913-918, 1992 8. Gleason-Morgan D, Church JA, Bagnall-Reeb H, et al: Complications of central venous catheters in pediatric patients with acquired immunodeficiency syndrome. Pediatr Infect Dis J 10:1114, 1991 9. Weiner ES, McGuire P, Stolar C, et al: The CCSG study of venous access devices: An analysis of insertions and causes for removal. J Pediatr Surg 27:155-164, 1992