P&T News: October 1995, Vol. 16, No. 4

Vancomycin Therapeutic Drug Monitoring in Adult Patients

Stephen C. Bergquist, M.S. and Michael B. Edmond, M.D.
Peer Review Status: Internally Peer Reviewed by Bradley E. Britigan, M.D., Professor and Director, Division of Infectious Diseases, Department of Internal Medicine

Currently, vancomycin is the second most frequently used antibiotic at UIHC. Due to this high use and its perceived potential for toxicity, monitoring of serum vancomycin levels in patients receiving this antibiotic has become the usual practice. Vancomycin serum assays were performed on 5,425 samples at UIHC during fiscal year 1994-95 and over 5.7 million assays were performed nationwide, even though this practice has not been shown to benefit patient care.2-4 This article will review the medical literature describing the use of serum vancomycin assays in therapeutic drug monitoring. It will also provide guidelines for the selective use of vancomycin serum assays to promote a more rational and cost-effective approach to their clinical use.

Drug Serum Level Monitoring
The use of drug serum level monitoring is clinically indicated if the following criteria are met:

1. A correlation between the serum drug concentration and clinical efficacy or toxicity can be defined;
2. A substantial variability in pharmacokinetics between patients exists;
3. Clinical efficacy or toxicity is difficult to measure or has a delayed onset; and
4. A sensitive assay is readily available.5

The long-term practice of drug serum level monitoring for agents such as aminoglycosides, digoxin, theophylline, antiarrhythmics, lithium, and anticonvulsants is well-founded clinically and is consistent with the above-mentioned criteria. Vancomycin generally meets Criteria 3 and 4, but does not generally satisfy Criteria 1; Criteria 2 is partially met. Nevertheless, a defined therapeutic range for vancomycin is frequently reported as 30 to 40 mcg/ml for the peak concentration and 5 to 10 mcg/ml for the trough concentration.6

Antimicrobial Activity Considerations
At UIHC vancomycin-susceptible staphylococci and streptococci typically exhibit a minimal inhibitory concentration (MIC) of 1 mcg/ml or less. Vancomycin only exhibits concentration-dependent killing of bacteria below concentrations of 1 mcg/ ml.78 It produces time-dependent killing above this concentration.98 If 50% protein binding of vancomycin is assumed and only free drug can interact with bacteria, it can be postulated that a trough serum level of 2 to 5 mcg//ml should be adequate to treat most vancomycin-susceptible organisms. Thus, the vancomycin serum level would be at or above the MIC for susceptible bacteria for the entire dosing interval. Most dosing methods using body weight, renal function and/or individualized pharmacokinetic dosing calculations produce trough serum levels at or above these ranges.

Clinical Efficacy and Serum Concentrations
Studies to date in the medical literature have not demonstrated a direct relationship between vancomycin serum levels and clinical response. A recent study by Zimmerman et al suggested that patients were more likely to become afebrile within 72 hours if peak concentrations were 20 mcg/ml or higher and trough concentrations were greater than 10 mcg/rnl.9 However, the study was a non-randomized, retrospective chart review.

Studies in adult and pediatric patient populations using fixed vancomycin doses, dosing nomograms, and doses individualized by pharmacokinetic calculations to produce peak and trough serum levels within a defined therapeutic range, demonstrate clinical effectiveness in the treatment of gram-positive infection. 10-16

Serum Level Toxicity Relationships
When examining the relationships between peak and trough vancomycin serum levels and the subsequent development of nephrotoxicity or ototoxicity, it is difficult to predict risk based on these data. While over 50 cases of ototoxicity have been attributed to vancomycin, it is difficult to directly implicate vancomycin as the cause due to the presence of other ototoxic drugs or disease conditions which also may have predisposed the patient(s) to ototoxicity. It is also difficult to define a range of peak or trough serum concentrations that correlate with an increased risk of ototoxicity. Many patients who experienced ototoxicity had serum levels that overlapped or were within the defined therapeutic range. 17

By 1987 over 167 cases of vancomycin-associated nephrotoxicity had been reported.4 The causative role of vancomycin is again unclear due to the presence of a number of confounding variables in these patients, including: therapy with other nephrotoxic drugs; prior antibiotic regimens; worsening heart failure; use of contrast dyes; hypotension, possibly secondary to rapid infusion of vancomycin; and incomplete patient data. The reported incidence of vancomycin-induced nephrotoxicity varies widely depending on the criteria used to define nephrotoxicity; it generally ranges between 0 to 35%.3 A study by Rybak et al that excluded other potential causes of renal dysfunction reported the incidence to be 5%;18 this incidence probably more accurately reflects the true incidence of vancomycin-induced nephrotoxicity.

The relationship between trough vancomycin serum levels and nephrotoxicity is also unclear. Some authors have found an increased incidence of nephrotoxicity with trough serum levels greater than 10 mcg/ml.'8'9 However, others have not confirmed these findings. A recent retrospective study in patients with gram-positive bacteremia demonstrated a mean vancomycin trough of 23.2 mcg/ml in patients with nephrotoxicity. The non-nephrotoxic patients had a mean trough serum level of 10.2 mcg/ml.9 These data seem to suggest a relationship between elevated trough serum levels and nephrotoxicity. However, due to the significant renal clearance of vancomycin, renal dysfunction would predispose patients to elevated serum vancomycin concentrations. Thus, it is difficult to conclusively link elevated serum vancomycin levels and nephrotoxicity.

Patient Factors Altering Vancomycin Pharmacokinetics
A number of patient-specific factors may alter the pharmacokinetic behavior of vancomycin thereby dictating a modification in dosing. Renal dysfunction, increasing age, morbid obesity, and severe burns have been demonstrated to alter the disposition and clearance of vancomycin.21-23 The clinical use of traditional hemodialysis, high-flux hemodialysis, or hemofiltration facilitates vancomycin clearance.2426 The monitoring of vancomycin serum levels in these clinical situations may assure effective therapy while minimizing the cost of unnecessary doses.

Proposed vancomycin Serum Level Monitoring Guidelines
Review of these clinical issues and the published data makes it apparent that a change in the approach to vancomycin serum level sampling should be implemented at the UIHC. Patients receiving vancomycin serum level monitoring should be carefully selected to maximize clinical benefit and minimize the waste of resources. Vancomycin serum level sampling guidelines should assist physicians and other health care professionals in this selection process and serve as an initial mechanism for the selection for patients in whom vancomycin serum level sampling might facilitate clinical management (Figure 1). However, the decision should also be based on sound clinical judgment, the patient's clinical response, and course of the infectious process being treated. When serum level monitoring is warranted, use of the guidelines provided in Figure 2 will facilitate the appropriate timing of serum level sampling, and hence their interpretability.

Figure 1. Patient Selection Guidelines for Vancomycin Serum Level Monitoring in Adults

Serious gram-positive infections such as sepsis, endocarditis or meningitis in which aggressive dosing and longer duration of therapy may be required. Patients with burns involving greater than 25 % of their body surface area. Patients in whom culture and susceptibility results indicate an intermediate MIC to vancomycin (MIC > 1 < 6 mcg/ml) for a susceptible bacterial isolate where higher daily vancomycin doses may be required. Patients with documented renal dysfunction, serum creatinine greater than 1.8 mg/dl, or who exhibit rapidly changing renal function but are not receiving peritoneal or hemodialysis. Patients undergoing traditional hemodialysis, high flux hemodialysis, or hemofiltration.

__________ Approved by the Pharmacy and Therapeutics Subcommittee, October 1995.

Review of Vancomycin Serum Level Sampling at UIHC
Using similar guidelines to those in Figure 1, a prospective review of adult patients receiving intravenous vancomycin was conducted during November, 1994 at UIHC. One hundred seventy-six adult patients received intravenous vancomycin. Sixty-five patients received vancomycin for greater than 48 hours. Eighty percent (52/65) of these patients had vancomycin serum levels drawn. Fifteen of 65 patients (23 %) met study criteria for vancomycin serum level sampling. Eighty-five of the 199 vancomycin levels drawn were consistent with the literature-based guidelines.

Forty-six percent (24/52) of the patients having serum level determinations had a subsequent vancomycin dosing change: 57% (14/24) of patients had dosing changes based entirely on the serum level result; 33% (7/24) of patients had dosage changes based on their clinical status; and 10% (3/24) of patients had a dosage change made due to both a combination of clinical response and the vancomycin serum level result.

The results of this prospective review can be summarized by the following:

1. A majority of patients (14/24 or 57 %) had vancomycin dosage changes based largely on the serum level result.
   
2. In the 30-day review period, 114 vancomycin levels were drawn in 176 adult patients that may not have been required in the management of the patient.

Vancomycin serum levels were frequently ordered empirically at the initiation of therapy independent of patient clinical response. Serum levels were also ordered in patients receiving vancomycin for prophylaxis. This review indicates the need for greater scrutiny in the selection of patients requiring vancomycin serum level sampling as a component of their clinical management.

Summary
This focused review has attempted to discuss areas of controversy associated with the clinical use of vancomycin serum level monitoring and provide guidelines for appropriate patient selection. At UIHC during fiscal year 1994-95, $286,000 in patient charges (5425 assays at $52.75 patient charge per assay) resulted from vancomycin serum assays. During our November 1994 review, 114 vancomycin serum levels were deemed via our review criteria to be unnecessary in the management of the patient. This could be extrapolated to approximately $72,000 in patient charges for unnecessary vancomycin serum levels for the previous fiscal year. These figures do not include other associated personnel costs incurred in the drawing of serum levels and interpreting this information.

For many patients, the need to monitor vancomycin levels is not well-supported by clinical studies of toxicity data. Vancomycin serum level sampling can be justified in certain patient populations or by clinical conditions such as renal dysfunction. The proposed guidelines may assist in the initial selection of patients for serum level sampling. The need to employ serum levels in all patients receiving vancomycin is unnecessary and a potential waste of costly resources.

Figure 2. Guidelines for Vancomycin Serum Level Sampling

How to Obtain Vancomycin Serum Levels Draw the trough level immediately prior to the dose Administer vancomycin over at least 60 minutes for doses < 1 gram. All doses 1 gram or more should be administered over at least 90 minutes. Flush the IV line after administration. Draw the peak level at least 60 minutes after the end of the infusion. If vancomycin is administered via a peripheral site, the serum level should be drawn from the opposite extremity. When to Obtain Vancomycin Concentrations Follow guidelines for vancomycin serum level sampling (see Figure 1). Obtain vancomycin serum levels at steady state conditions (3 to 5 half-lives or after 3 to 5 maintenance doses). Random levels may be drawn in renal failure patients to assess time for redosing or effects of dialysis procedures, but should generally not be performed on a daily basis. Patients with normal renal function requiring long-term therapy (i.e., 4 to 6 weeks or longer) can be monitored using weekly trough serum levels only. Patients receiving vancomycin for short-term prophylaxis (48 hours or less) do not require serum levels.

__________ Approved by the Pharmacy and Therapeutics Subcommittee. October 1995.

References

1. Mayo Clin Proc 1991; 66:1165-70.
2. Ann Pharmacother 1993; 27:594-98.
3. CID 1994; 18:533-40.
4. Clin Pharm 1987; 6:652-54.
5. Advances in therapeutic drug monitoring. Proceedings of the First International Congress of Therapeutic Drug Monitoring. Tokyo :Enterprise. 1990: 297-302.
6. Mayo Clin Proc 1977; 52:631-4.
7. Antimicrob Agents Chemother 1990; 34:1869-74.
8. Antimicrob Agents Chemother 1992; 36:1766-9.
9. Pharmacotherapy 1995; 15:85 91.
10. J Pediatr 1980; 96:119-26.
11. Ann Intern Med 1982; 97:344-50.
12. Antimicrob Agents Chemother 1991; 35:2467-72.
13. J Antimicrob Chemother 1982; 9:69-74.
14. Ann Intern Med 1981; 94: 343-6.
15. Clin Pharm 1987; 6:795-9.
16. Clin Pharm 1991; 10:129-32.
17. Antimicrob Agents Chemother 1983; 23:13841.
18. J Antimicrob Chemother 1990; 25:679-87.
19. Am J Med 1987; 83:1091-7.
20. Arch Intern Med 1989; 149:1777-81.
21. J Antimicrob Chemother 1984 (Suppl D); 14:43-52.
22. Clin Pharmacol Ther 1984; 36:803-10.
23. Clin Pharmacol Ther 1988; 44:9-13.
24. Clin Nephrol 1983; 20:85-8.
25. Clin Pharmacol Ther 1986; 40:425-30.
26. Ann Pharmacother 1994; 28:512-14.

Adverse Drug Reaction Report

Ticlopidine-Induced Hematologic Toxicity
Ticlopidine (Ticlid(R)) is an antiplatelet drug that acts by decreasing platelet aggregation and circulating platelet aggregates, and by prolonging bleeding time. It is indicated to reduce the risk of thrombotic stroke in patients who have experienced stroke precursors, and in patients who have had a completed thrombotic stroke. The manufacturer states that use of ticlopidine should be reserved for those patients intolerant to aspirin therapy because of the associated risk of neutropenia and agranulocytosis. There have been recent reports of severe hematologic adverse events secondary to ticlopidine use.

A 62-year-old man was admitted to the hospital with complaints of tiredness, abdominal pain, and nausea. Among the medications the patient was taking for treatment of diabetes and peripheral arteriopathy was ticlopidine 500 mg a day for 16 weeks. Routine clinical test uncovered neutropenia, as well as an inflammatory syndrome. Ticlopidine was discontinued and the hematologic abnormalities resolved in one month.

Another case report involved a 51-year-old man whose admission to the hospital was precipitated by fever and a four-week history of fatigue. For the previous ten weeks, the patient had been taking ticlopidine 500 mg a day for treatment of iterative transient cerebral attacks. Bone marrow biopsy and other hematologic tests were consistent with agranulocytosis and marked erythroid hypoplasia of the megakaryocyte type. The drug was discontinued and the patient's hematologic parameters returned to normal.

A 69-year-old man had been taking ticlopidine for just under eight weeks for a cerebro-vascular stroke and arterial hypertension. He was admitted to the hospital with a seven-day history of fatigue and several purpuric skin lesions on the trunk. A bone marrow aspirate along with other tests demonstrated aplastic anemia. After discontinuation of ticlopidine and conservative treatment, the patient's hematologic parameters gradually improved over the next several weeks. However, his bone marrow aspirate remained hypocellular with an absence of megakaryocytes and he continued to receive platelet transfusions.

__________

Abstracted from: Lesesve JF et al. Hematological Toxicities of Ticlopidine. Am J Hematol. 1994;45:149-50.

Over the past 18 months, there have been two patients with ticlopidine-induced hematological adverse effects admitted to UIHC. The first case involved a 57-year-old man, with cerebrovascular disease, who had taken ticlopidine 500 mg a day for one month. The drug was discontinued and the next day he was transferred to the UIHC with fever, left leg weakness, and a complaint of confusion and somnolence. He was diagnosed with neutropenia and agranulocytosis. Within three days of admission, his neutrophil count returned to normal and the patient remained afebrile throughout his hospital stay.

A 58-year-old woman experienced a presumed transient ischemic attack and was placed on ticlopidine 500 mg a day. One week later, a CBC revealed no abnormalities. Approximately three weeks after beginning ticlopidine, the patient began to "feel sick" with complaints of fever, nausea, decreased appetite, easy bruising, petechiae on her arms, and reddish-brown urine. Her complaints continued to escalate for the next few days until her local physician sent her to UIHC. Ticlopidine was discontinued the day of admission. Clinical tests, including a bone marrow biopsy, revealed the patient had thrombotic thrombocytopenic purpura. She was treated with plasmapheresis and platelet transfusions. Within nine days after her admission, her blood counts stabilized. She was discharged after 11 days.

Prescribing information for ticlopidine states that the incidence of neutropenia is 2.4% and agranulocytosis is 0.8% in patients who receive the drug. Thrombocytopenia, alone or together with neutropenia, has been rarely reported. Cases of pancytopenia and thrombotic thrombocytopenic purpura, some of which have been fatal, have also been rarely reported. Although not specifically mentioned in the product information, there have been several cases of aplastic anemia secondary to ticlopidine use reported in the literature. *

As instructed by the manufacturer, rigorous hematologic monitoring must be conducted when patients are prescribed ticlopidine. Complete blood counts and white cell differential counts should be performed every two weeks starting from the second week after initiation of therapy to the end of the third month of therapy. After the first three months of therapy, CBCs only need to be repeated in patients with signs or symptoms which suggest infection, neutropenia, or thrombocytopenia .

* References available upon request from the Drug Information Center .

1. Documented or suspected gram-negative infection in a patient with a type I (i.e., anaphylactoid) hypersensitivity or other unacceptable allergic reactions (e.g., severe skin reaction) which precludes the use of comparable penicillins and/or cephalosporins (e.g., cefotaxime, ceftriaxone, piperacillin).
   
2. Formal Infectious Diseases Consult recommendation.

It is important to remember that aztreonam:

• Should NOT be used as a substitute for an aminoglycoside when predictable additive or synergistic activity is desired (e.g., gram-negative aerobic rods and/or enterococci).
• Exhibits NO predictable additive or synergistic effects when used in combination with other beta-lactam agents.
• Has NO gram-positive or anaerobic activity.
• Has NO role in the therapy of systemic enterococcal infections when used alone or in combination with other antibiotics.

As with other protocol drugs, the prescriber must complete a UIHC Protocol Antibiotic Order Form specifying the indication for use of aztreonam. In the absence of a completed order form, the order for aztreonam will be discontinued and the prescriber notified as per Pharmacy and Therapeutics Subcommittee policy.

Next Page | Title Page