Editorial
May 2001
Volume 96, Number 5
Pages 1327-1329

Controlling Pain in Liver Biopsy, or "We Will Probably Need to Repeat the Biopsy in a Year or Two to Assess the Response"

Stephen H. Caldwell, M.D

Liver biopsy is a fundamentally important procedure in the evaluation of chronic and acute liver diseases. Although specific figures are not available, the epidemic of hepatitis C, the growing number of therapeutic options in various liver diseases, and, more recently, the recognition of the potential severity of fatty liver have increased the number of biopsies in most centers over the past 10 yr. Based on the experience at our center, where approximately five biopsies are performed per week, it can be conservatively estimated that over 30,000 liver biopsies are performed annually in the US (assuming an average of three similar centers per state). Surrogate markers for measuring the degree of liver injury continue to attract attention but are unlikely to supplant histological examination in the foreseeable future (1). Because of the central role of liver biopsy in diagnosis, staging, and assessing treatment response, it is important to examine ways to make the procedure more acceptable to patients. Castéra et al. have addressed this problem in an article in this issue (2) reporting on a placebo-controlled trial of self-administered nitrous oxide in patients undergoing percutaneous liver biopsy.

Postbiopsy pain is the most common complication of liver biopsy, although the incidence varies depending on the assessment of severity (3). Moderate to severe pain, often requiring hospitalization, is seen in 1-5% of patients in past series (4, 5). Aside from trauma to adjacent organs, the pain in this situation is more often associated with the need for blood transfusions and hence is likely related to frank bleeding, although most recover with only supportive measures. The higher range is seen with cutting needles (6) (such as Tru-Cut [Travenol, Deerfield, IL]) as opposed to aspiration needles [such as Menghini style (7) needles], although this is disputed (5). Similarly, the amount of moderate to severe pain experienced by the patient varies with the experience of the practitioner (8), although there is also disagreement on this point (5).

The incidence of less severe pain has been reported in several studies. Pain requiring postprocedure analgesia is seen in approximately 50% of patients undergoing biopsy with an automated cutting needle, but this figure is reduced to approximately 30% when ultrasound is used to guide the cutting needle (9). In contrast, in a series of 101 consecutive patients undergoing non-ultrasound guided biopsy with an aspiration needle (1.6 mm, Jamshidi, Allegiance, McGaw Park, IL), 34% of the patients experienced pain requiring analgesics (10). In this issue's Castéra et al. study, which also used an aspiration needle (1.8 mm, Hepafix, Braun, Melsungen, Germany), 39% of patients in the placebo group reported postprocedure pain. Thus, the overall frequency of pain requiring analgesics appears to be around 30-50% in patients not receiving prebiopsy medications. It is worth noting that moderate pain was reported in about 30-40% of patients using the non-ultrasound guided aspiration needles (Menghini style, such as Jamshidi or Hepafix) in two studies. A similar figure for the cutting needle was obtained only with ultrasound guidance. It is speculative, but this difference may relate to the loss of tactile guidance with the automated cutting needles relative to the aspiration needles and perhaps the longer dwell-time of the latter needle style (11).

Outside of frank trauma to adjacent organs or severe bleeding, the mechanism of postbiopsy pain is uncertain. Small hematoma formation has been implicated in one study that reported that 20% of liver biopsy patients had postbiopsy hematomas (12). However, these findings were disputed in a number of letters written in response to that report (13, 14, 15). On the other hand, surface bleeding at the biopsy site is almost universal, as any laparoscopist will attest. Oozing typically commences immediately when the needle is withdrawn and sometimes begins with a slight gush. Liver bleeding time—the time until the oozing spontaneously stops—is typically 3-5 min (16). The amount of blood is usually only 30-50 ml as estimated by gross inspection. However, it seems likely that this hemodynamically inconsequential amount of blood is the source of most postbiopsy pain because of irritation of the capsule and peritoneum. This relationship is supported by the low frequency of pain (10%) reported in a study of percutaneous biopsy site plugging (17) and the significantly decreased liver bleeding time noted laparoscopically when plugging agents are used (18).

Castéra and colleagues report a novel, convenient, inexpensive means of controlling pain in patients undergoing standard percutaneous liver biopsy. One hundred average risk patients undergoing percutaneous biopsy with an aspiration style needle (1.8 mm, Hepafix) with local xylocaine were randomized to either a mixture of equal parts N₂O/oxygen or an oxygen placebo. The study was blinded and pain was measured by a previously reported visual analog pain score in which the patient made a mark on a 10-mm line to indicate no pain (at the far left of the line) or maximal pain (at the far right). The gas was delivered via face mask by a self-activated device that required triggering by the patient to activate. When drowsiness developed, the patient relaxed the grip on the mask and thus turned off gas flow. Using this device, the authors noted a significant decrease in the mean reported pain scores and a complete absence of pain in 19 patients (38%) in the treatment group versus only two (4%) in the placebo group. One patient (2%) in the treatment group experienced severe pain, compared to nine (18%) in the placebo group. A nurse, trained in use of the dispenser, supervised the administration of the nitrous oxide, which cost an estimated $4/patient. Their results offer potential benefit in pain relief, as compared with results using either an aspiration or a cutting needle in nonpremedicated patients.

In the US, the performance of percutaneous liver biopsy has undergone remarkable changes recently. Although biopsy remains a staple in most academic GI/hepatology units, the procedure appears to be performed less commonly by gastroenterologists in private practice. We recently polled 16 of our past fellows who are now in private practice. Only one-half continued to perform liver biopsy on a routine basis, and one-half did not perform the procedure at all. Relatively poor reimbursement in proportion to liability for a high risk procedure was cited as a major reason for not performing biopsy. Although abundant historical and more recent literature support the use of traditional percussion-guided biopsy (19, 20), heightened worry over liability using this method as opposed to ultrasound-guided biopsy has added to a shift toward performance of biopsy in the radiology suite. This carries with it a move toward more frequent use of automated cutting needles and a rate of postprocedure pain similar to that seen with percussion-guided aspiration needles (around 30% requiring postprocedure analgesia).

Although the risk of a serious problem is low, the possibility of a severe adverse event and the uncertainty of interpreting postprocedure pain can cause a great deal of angst. In comparison to the utility of the procedure in liver disease and the seriousness with which one must approach liver biopsy, the typical reimbursement seems remarkably casual. The relative value unit for standard liver biopsy was 3.43 last year (Current Procedural Terminology code 47000). In our area, this translates to a Medicare allowable reimbursement of $106 for the professional fee. For a procedure with potentially severe complications, this amounts to an inadequate reimbursement for many practitioners facing tight schedules and decreased reimbursements in other areas. Some gastroenterologists have adapted to this climate by incorporating ultrasound guidance into their own practice, although the Medicare allowable for guidance is a modest $34 in our region, whereas Blue Shield allows $45. In comparison, the Medicare allowable for a skin biopsy is $41.40 and that for percutaneous kidney biopsy is $160 in our region (not calculated with ultrasound guidance).

Is use of nitrous oxide likely to be adopted in endoscopy or radiology units performing liver biopsy? Castéra et al. have demonstrated the efficacy of their nitrous oxide system in controlling pain after routine percutaneous biopsy in this randomized, placebo-controlled trial (2). However, adoption of the system poses several problems. First, as pointed out by the authors, the system costs about $700 to set up (including the delivery apparatus with valves and mask). In addition, there are concerns about possible exposure to personnel administering the agent. Moreover, nurse training would be required to safely use this form of analgesia. Nursing supervision of conscious sedation is standard in most endoscopy units but requires additional arrangements in many radiology suites where sedation and analgesic use has not been standard practice. Adoption of this technique is furthermore unlikely to avoid the need for an i.v. line placed as a safety measure in case of a complication. In addition, many practitioners have already adopted routine use of premedications before the biopsy.

Thus, there are a number of obstacles that make widespread adoption of this system unlikely in the US. On the other hand, the authors have highlighted the need for attention to patient comfort in this increasingly common procedure, and they have provided evidence supporting a valid means of substantially achieving this goal. Their findings should be borne in mind the next time one finds oneself discussing liver biopsy with a patient and the potential need for repeating the biopsy at some point in the future. It is unclear whether or not similar results in pain control could be obtained with more routine use of prebiopsy i.v. sedation or small doses of i.v. analgesics. However, use of a self-activated nitrous oxide analgesic system would likely be welcome by most patients and appears to offer effective analgesia at a low price per patient and with a quick recovery.

Acknowledgments

The author thanks Ms. Greta Hedberg for her assistance in obtaining reimbursement information.

Division of Gastroenterology and Hepatology, Department of Internal Medicine, University of Virginia, Charlottesville, Virginia

References

1. Rosenberg W, Burt A, Becka M, et al. Automated assays of serum markers of liver fibrosis predict histological hepatic fibrosis. Hepatology 2000;32:183A.

2. Castéra L, Negre I, Samii K, Buffet C. Patient-administered nitrous oxide/oxygen inhalation provides safe and effective analgesia for percutaneous liver biopsy: A randomized placebo-controlled trial. Am J Gastroenterol 2001;96:1553-7.

3. Grant A, Neuberger J. Guidelines on the use of liver biopsy in clinical practice. Gut 1999;45(suppl IV):IV1-11.

4. Janes CH, Lindor KD. Outcome of patients hospitalized for complications after outpatient liver biopsy. Ann Intern Med 1993;118:96-8.

5. Perrault J, McGill DB, Ott BJ, Taylor WF. Liver biopsy: Complications in 1000 inpatients and outpatients. Gastroenterology 1978;74:103-6.

6. Piccinino F, Sagnelli E, Pasquale G, Giusti G. Complication following percutaneous liver biopsy. A multicenter retrospective study on 68,276 biopsies. J Hepatol 1986;2:165-73.

7. Menghini G. One-second needle biopsy of the liver. Gastroenterology 1958;35:190-9.

8. Froehlich F, Lamy O, Fried M, Gonvers JJ. Practice and complications of liver biopsy—results of a nationwide survey in Switzerland. Dig Dis Sci 1993;38:1480-4.

9. Lindor KD, Bru C, Jorgensen RA, et al. The role of ultrasonography and automatic-needle biopsy in outpatient percutaneous liver biopsy. Hepatology 1996;23:1079-83.

10. Hespenheide EE, Caldwell SH, Dye KR, et al. Pain and analgesic use following percutaneous liver biopsy. Hepatology 1998;28:768A.

11. Garcia-Tsao G, Boyer JL. Outpatient liver biopsy: How safe is it? Ann Intern Med 1993;118:150-3.

12. Minuk GY, Sutherland LR, Wiseman DA, et al. Prospective study of the incidence of ultrasound-detected intrahepatic and subcapsular hematomas in patients randomized to 6 or 24 hours of bed rest after percutaneous liver biopsy. Gastroenterology 1987;92:290-3.

13. Pelckmans PA, Pen JH, Michielsen PP, Van Maercke YM. Hematomas after percutaneous liver biopsy. Gastroenterology 1988;94:249 (letter).

14. Spinzi GC, Terruzzi V, Minoli G. Hematomas after percutaneous liver biopsy. Gastroenterology 1988;94:249 (letter).

15. D'Aquino M, Michieletto L, Caprioglio L. Hematomas after percutaneous liver biopsy. Gastroenterology 1988;95:575 (letter).

16. Ewe K. Bleeding after liver biopsy does not correlate with indices of peripheral coagulation. Dig Dis Sci 1981;26:388-93.

17. Tobin MV, Gilmore IT. Plugged liver biopsy in patients with impaired coagulation. Dig Dis Sci 1989;34:13-5.

18. Falstrom JK, Goodman NC, Moore MM, et al. Use of fibrin sealant as a hemostatic plug for percutaneous liver biopsy tract. J Vasc Invasive Radiol 1999;10:457-62.

19. Stotland BR, Lichtenstein GR. Liver biopsy complications and routine ultrasound. Am J Gastroenterol 1996;91:1295-6.

20. Smith CJ, Grau JE. The effect of ultrasonography on the performance of routine outpatient liver biopsy. Hepatology 1995;22:384A.

Reprint requests and correspondence: Stephen H. Caldwell, M.D., Division of Gastroenterology and Hepatology, Box 10013, University of Virginia Health System, Charlottesville, VA 22906-0013.