A Prospective Evaluation of Elevated Serum Theophylline Concentrations to Determine If High Concentrations Are Predictable PAULA. GREENBERGER, M.D., JAMES A. CRANBERG, M.D., MICHAEL A. GANZ, M.D., GARY L. HUBLER, M.S., R.Ph., Chicago, Illinois
PURPOSE: To evaluate
prospectively whether serum theophylline concentrations of 25 mg/L and greater were predictable (and presumably preventable) by use of basic pharmacohinetic WlCulatiOM.
DEWGN: Prospective PATENTS Fifty-five
study. patients with a serum theophylline concentration of at least 25.0 n&L were evaluated initially and if subsequent elevated theophylline concentrations occurr’ed. INTERVENTIONS: The predicted steady-state serum theophylline concentration was calculated from the dosage rate divided by the predicted clearance to determine how many elevated concentrations (greater than 20 mg/L) were predictable. Predicted clearances were 0.04 L/l&hour for normal subjects less than 70 years of age and 0.02 L/hg/hour for patients with congestive heart failure, chronic obstructive pulmonary disease, or liver disease. Esthnated clearances were determined and compared with predicted clearances. If patients did not have steady-state concentrations, additional calculations were made. MAIN RESULTS: Prom 6968 consecutive theophylliue determinations, 69 (1.08 % ) samples from 55 patients were 25 mg/L or higher. Predictably high concentrations occurred in 23 of 33 (69.7%) fully evaluable cases. These concentrations occurred because of a failure to consider decreased elimination clearance from congestive heart failure, chronic obstructive pulmonary disease, or hepatic disease. Five fatalities occurred, and iu two cases, theophylline appeared to contribute to the patient’speak Three other patients experienced syncope. The predicted elimination clearance of theophylline of 0.02 From the Section of Allergy-Immunology, Department of Medicine, Northwestern University Medical School, Chicago, Illinois. This work was supported by United States Public Health Service Grant Al 11403 and the Ernest S. Bazley Grant. Requests for reprints should be addressed to Paul A. Greenberger, M.D., Northwestern University Medical School, 303 East Chicago Avenue, Chicago, Illinois 60611-3010. Manuscript submitted January 4.1991, and accepted March 28,199l.
L/kg/hour was too high in eight patients over 70 years old with cardiac or pulmonary disease. Nursing and pharmacy oversights were identified as three patients were given two theophylline products simultaneously. CONCLUSIONS: Most elevated theophylline concentrations are predictable (and preventable) by basic pharmacohinetic calculations. Patients experiencing elevated theophylline concentrations often had comorbid conditions and were greater thau 60 years of age. The dosage rate of theophylline (n&hour) can be estimated from predicted clearance (L/l&hour) times desired steady-state serum concentration (n&L).
T
heophylline is a bronchodilator that has been recognized to be effective in the therapy of acute and especially chronic asthma and is known to have respiratory stimulatory properties as well. Although the therapeutic response to theophylline increases proportionately with the log of the serum concentration, the risk of toxicity also increases with higher concentrations, creating a relatively narrow therapeutic range for the drug [1,2]. Potential adverse effects from theophylline are well recognized and range from minor central nervous system stimulation and gastrointestinal symptoms such as anorexia, abdominal discomfort, nausea, and vomiting to ventricular arrhythmias and seizures [2]. As the greatest therapeutic response is achieved with serum theophylline concentrations greater than 10 mg/L, and serious adverse effects are most likely to be observed at concentrations greater than 20 mg/L, 10 to 20 mg/L is the generally accepted optimal therapeutic range for the drug
WI.
Serum theophylline concentrations may be estimated by knowing the patient’s weight and theophylline dosage, and by awareness of concomitant conditions or medications that are known to affect theophylline clearance [5-71. Inappropriately elevated concentrations are undesirable and pose a risk of theophylline toxicity to the patient, although some patients are asymptomatic despite substanJuly 1991 The American Journal of Medicine
Volume 91
67
tial elevations of serum theophylline concentrations. Elevated serum theophylline concentrations can occur for many reasons, which may include patient, nursing, pharmacy, and/or physician errors [8-121, which suggests that they may be preventable. We undertook a prospective evaluation of elevated serum theophylline concentrations in an attempt to determine the incidence and causes of elevated concentrations at Northwestern Memorial Hospital, a large teaching hospital. Specific attention was directed at whether serum theophylline concentrations greater than 25 mg/L were predictable (and presumably preventable) or nonpredictable using basic pharmacokinetic calculations.
PATIENTS AND METHODS We prospectively identified all patients who had serum theophylline concentrations exceeding 25 mg/L reported at the Northwestern Memorial Hospital Clinical Pharmacology Laboratory between September 1,1988, and August 30,199O. All assays were reported by the Clinical Pharmacology Laboratory and were performed using a fluorescence polarization immunoassay with an interassay coefficient of variation of less than 5%. The reported concentrations were obtained from patients from all areas of the hospital, including inpatient wards, outpatient clinics, and the emergency room. Cases were evaluated by review of the patient records, and in some cases by additionally interviewing the patient, on the same day that the elevated concentration was reported or shortly thereafter. The investigation was not announced to medical, nursing, or pharmacy staff but was known only to investigators and members of the Pharmacy and Therapeutics Committee, Northwestern Memorial Hospital, Chicago, Illinois. Salient features that were reviewed and recorded for each case included (1) the time and date that the elevated concentration was obtained as well as its temporal relation to the last dose of theophylline, (2) the theophylline formulation and dose being taken, (3) the length of time that the patient had been taking that dose, (4) concurrent medications, (5) patient characteristics including age, height, weight, smoking status, current and previous medical problems or illnesses, reason for current hospitalization, or current clinical status, (6) the presence or absence of signs or symptoms of theophylline toxicity, and (7) previous serum theophylline concentrations and dosages. Basic pharmacokinetic analysis was performed in each patient, including (1) prediction of theophylline clearance, (2) estimation of the actual clearance in each patient, and (3) prediction of steady-state
68
July 1991
The American Journal of Medicine
Volume 91
serum theophylline tered dose.
concentration
at the adminis-
Prediction of Theophylline Clearance The clearance was predicted by knowing the patient’s weight and using previously reported average theophylline clearances that are associated with certain patient characteristics. For patients with congestive heart failure, chronic obstructive pulmonary disease, liver disease, viral upper respiratory illness, or the use of agents such as cimetidine, erythromycin, and allopurinol, an average clearance of 0.02 L/kg/hour was assumed [3]. For healthy, nonsmoking adult patients less than 70 years of age, an average clearance of 0.04 L/kg/hour was assumed. For otherwise healthy, nonsmoking adults of 70 years and older, a predicted clearance of 0.02 L/kg/hour was used [13]. For patients in whom factors that are known to be associated with increased theophylline clearance are present, such as smoking together with age less than 40 years or the use of drugs such as phenobarbital, phenytoin, or rifampin, an average clearance of 0.07 L/kg/hour was assumed [3]. Estimation of Patient Theophylline Clearance and Prediction of Serum Concentrations The patients’ estimated theophylline clearances were determined using the formula: elimination clearance (Cl) = theophylline dosage rate/serum concentration (C,,) at steady state [5]. These concentrations were considered to be at steady state with regard to theophylline pharmacokinetics (1) if medical records revealed no interruption of maintenance theophylline therapy or (2) if theophylline was initiated or the dosage was changed, and the drug had been administered for a minimum of four half-lives based on the predicted clearance for each patient. We used the formula ti/z = 0.693 V&l, where Vn is the volume of distribution calculated from 0.45 L/kg. In non-steady-state cases, the concentration of theophylline at time in hours was estimated from the following equation: C, =
loading dose v eekt D +
infusio;
rate (1 _ e-kt)
When sequential sera were obtained after the infusion was stopped, the elimination rate constant k was determined when we estimated t1/2. When additional sera were not obtained or were not helpful, we calculated k = Cl/Vn using the predicted clearance.
ELEVATED
After comparing the estimated theophylline clearancewith the initial prediction basedon literature values, we estimated the steady-stateserum theophylline concentration by the formula C,, = theophylline dosage rate (mg/hour)/estimated clearance(L/hour). When steady-stateconditions were not present,we estimated appropriatenessof dosagesand estimated serum theophylline concentrations such as with the formula: A concentration = A theophylline dose/volumeof distribution if a loading dosehad been given. If the drug had been discontinued,after apparent steadystate had been reached,we attempted to determine the half-life of theophylline if subsequentserum sampleswereobtained. Suchestimateswerecomparedwith a calculated Ti/2 of (0.693X Vn)/Cl. We consideredthe reported elevatedconcentration predictable (and thus potentially preventable) when basic pharmacokinetic calculations resulted in expected theophylline concentrations greater than 20 mg/L. Unfortunately, of 55 patientswho had serumtheophylline concentrationsof 25 mg/L or greater,13 patients are not included in the pharmacokinetic analysis because necessary clinical information (weight, theophylline preparation, dosage,and so forth) was unavailable and unretrievable. The mean theophylline concentration in those 13 patients was 28.9f 4.2 mg/L (n = 14).Five of 14sera were from the emergency room or ambulatory clinics. RESULTS Of 6,368theophylline concentrationsreported at NorthwesternMemorial Hospital betweenSeptember 1,1988,and August 30,1990,69determinations (1.08%)were greater than 25 mg/L in 55 patients. The mean of the 69 elevated serum theophylline concentrationswas 29.0 f 3.9 mg/L. Five patients (9.1%)died; in two cases(Patients 7 and 35), theophylline toxicity probably contributed to the patient’s death. Three other patients (5.5%)experienced syncopal episodes.One patient presented with a grand mal seizure.Just two elevatedconcentrations were from a recognizedsuicidal gestureor attempt. Pertinent features of eachcaseare listed in Tables I and II, including patient age,predicted and measuredserum theophylline concentrations, theophylline preparation, dosebeing taken at the time the elevated concentration was discovered, predicted and estimated theophylline clearance data, and whether there wasevidenceof toxicity to the patient from theophylline. The elevatedserum theophylline concentrationswere believedto have beenpredictable in 23 (69.7%)of the 33 patients. In
SERUM
THEOPHYLLINE
CONCENTRATIONS
/ GREENBERGER
ET AL
10 patients in whom adequatedata were obtained, our estimatesdid not predict a serum theophylline concentration of more than 20.0 mg/L. We were unable to predict in nine other casesbecauseof partial data such as absenceof weight or dosages, which limited assessment. Predictable Elevations of Theophylline
Theophylline concentrationswere elevated predictably in 23 patients. The major explanation for such elevated concentrations was that the prescribeddoseof theophylline wasexcessivein view of decreasedtheophylline clearance.In most cases, the clearancewasbelievedto be abnormal because of the presenceof congestiveheart failure and/or chronic obstructive pulmonary disease.Patient 1 had an elevatedtheophylline concentrationthat superficially would not have been predictable based on theophylline clearancesthat were calculated from concentrationspreviously obtained from this patient. However,at the time of the initial elevated concentration of 31.8mg/L, the patient had been admitted for an exacerbationof congestiveheart failure and chronic obstructive pulmonary disease. His rapidly deterioratingclinical condition had not led to an associatedreduction of the theophylline dosage.Sevenelevatedtheophylline concentrations wereobtainedovera period of 2 months subsequent to the initially elevated concentration during admissions for recurrent exacerbationsof the aforementioned medical problems.Each of thesesubsequent concentrationswaselevatedpredictably. For Patient 1, the highest serum theophylline concentration was44.8mg/L and clinically wasassociated with a syncopal episode. Patients 3,4, and 11wereinadvertently administered two forms of theophylline. Four patients (Patients 12, 30, 34, and 37) ingested theophylline in greaterthan authorized doses. Nonpredictable
Elevations of Theophylline
Elevated concentrationsobtained in 10 patients werebelievednot to havebeenpredictablebasedon predicted theophylline clearances.In nine additional patients, incompletedata made a meaningful estimate impossible. The high serum theophylline concentrationsin Patients 22 and 24 occurredas a result of suicidegesturesor attempts. Patient 6 had a history of chronic obstructive pulmonary disease and cardiacarrhythmias, and was admitted after a syncopalepisodewith shockand respiratoryfailure. A reducedtheophylline clearancefrom her compromised cardiac and pulmonary status likely explained the elevated serum concentration of 26.9 mg/L and estimated clearanceof 0.013L/kg/hour.
July
1991
The American
Journal
of Medicine
Volume
91
69
ELEVATED
SERUM
THEOPHYLLINE
CONCENTRATIONS
/ GREENBERGER
ET AL
TABLE I !Mxted Pharmacokinetic Parametersof 42 Patients with Serum Theophylline ConcentrationsGreater Than 25 mg/L Measured Theophylline Concentration*
Predicted Theophylline Concentration
Predicted Theophylline Clearance
Estimated Theophylline Clearance
(mg/Ll
(mgll)
Wk@r)
kwllr)
Theo-Dur 600 mg tid
31.8
39.2
0.020
0.024
Theo-Dur 300 Aminophylline Theo-Dur 300 Choledyl 200
25.6 32.5
20.4 31.3
0.040 0.020
0.028 0.019
27.5
23.8
0.020
0.017
30.6 26.9 29.8
16.7 17.2 28.8
0.040 0.020 0.020
0.022 0.013 0.019
mg tid mg tid 40 drip mg/hr mg x 1
30.6 25.8 25.0 27.9
35.7 24.4 21.6 17.8
0.020 0.020 0.040
0.022 0.019 0.028
Theo-Dur 300 mg qid Aminophyllinedrip 30 mg/hr
26.0 25.5
22.2 25.0
0.020 0.040
0.017 0.039
Theo-Dur 300 mg bid
26.3 25.0
21.5 28.4
0.020
0.015 0.022
Theo-Dur 300 mg tid bid
25.8 33.7
la.2 38.0
0.020
0.014 0.022
Theo-Dur 300 mg bid
25.7 32.0
20.5 32.0
0.020
0.016 0.020
Aminophylline 35 290mg/hr mg loading dose, 29 mg/hr infusion
25.3 25.7
20.3 19.6
0.020
0.016 0.015
Theo-Dur Unknown 300 number mg ofbidtablets
28.9 30.4
Unknown 21.0 dose
0.020 0.040
0.015
Unknown numberof 60 mg/hr tablets
28.9 28.1
Unknown 26.7 dose
0.020 0.040
0.020
&Q;;;ylline 500 mg tid Theo-Dur
44.9 39.4
Unknown 54.8 dose
0.020
0.023
1
Theo-Dur Unknown 200 mg bid
31.0 25.2
Unknown 14.4 dose
0.020
060
Product Dosage
mg bid drip 25 mg/hr mg bid mg bid
Theo-Dur 200 300 mg tid bid Theo-Dur 300 mg tid Theo-Dur 200 250 Theo-Dur 300 Aminophylline Theo-Dur 300
ZJT
Theo-Dur 300 200 mg tid bid
27.9 32.3
46.8 15.7
0.020
oiio
;i ii;
Aminophylline Theo-Dur 300 35 bid mg mg/hr Am;eph;;F 30 mglhr surreptitious
31.3 25.8 31.4
25.7 40.5
0.020 0.020
0.016 -
2
7:
Aminophylline 25 30 mglhr mg/hr
29.9 37.6
20.0 la.5
0.020
0.012 0.013
z; iz
2 17
0.040
0.040
0.026 -
59 :z
29.1 26.8 28.6 27.8 25.9 32.6
Unknown 11.0 Unknown Unknown
i:
Theo-Dur dose/unknown 800 mg/day Theo-Dur 300 mg tid Theo-Dur dose/unknown Theo-Dur 200 30 mg bid Aminophylline mglhr
1;::
0.020 0.040
-
0.037
Mean 28.4 f 6.5 tid = three times per day; bid = hvo times per day; qid = four times per day. *The initial theophyilineconcentration greater than 25 mgllfor each patient is presented.
Our predictedclearancesof either 0.04L/kg/hour or 0.02L/kg/hour exceededestimated clearances(Table I) for eight additional patients (Patients5,10,17, 20, 29, 31, 36, and 41). Patient 41 was apparently indiscreet in terms of theophylline use. Her usual dose was Theo-Dur@200 mg twice daily, which would result in a steady-stateserum concentration of 7.8 mg/L. The concentrationon admissionwas 25.9mg/L. Treatment with theophyllinewasdiscontinued andthen later resumed.At 200mg twice daily when steadystate occurred,her serumtheophylline concentrationwas 8.3 mg/L. We believethis caseis
70
July
1991 The American
Journal of Medicine
Volume 91
consistent with unauthorized self-overmedication with theophyllinein a patient with increasedwheezing during an asthma exacerbation. Manifestations
of Theophylline
Toxicity
Evidence of theophylline toxicity as noted in 25 (59.5%)of the 42 patients (Table II) consistedof nausea,jitterness, palpitations, tachycardia, arrhythmias, grandmal seizure,and syncope.Patient 11,who had a history of atrial fibrillation, presented with atrial fibrillation and rapid ventricular response.An associationof atrial fibrillation with the-
ELEVATED
SERUM
THEOPHYLLINE
CONCENTRATIONS
/ GREENBERGER
ET AL
TABLE II Initial Theophylline Concentrationsin 42 Patients and Potential Theophylline Toxicity
Patient Number
-
Age
Measured Theophylline Concentration (mglL)
Potential Toxkieb;t ldentiiication
Predictable Concentrations Greaterthan 20 m%L
31.8
None
Yes
32.5
Yes
Yes
25.6 27.5 30.6 26.9 29.8
Yes None Yes Possible Yes
Yes Yes No Unknown Yes
30.6 25.8 25.0 27.9 25.5 26.0 25.0 26.3 33.7 25.8 32.0 25.7 25.7 25.3 30.4 28.9 28.1 28.9 39.4 44.9 25.2 31.0 27.9 32.3 31.3 25.8 31.4 37.6 29.9 29.1 26.8 28.6 27.8 25.9 32.6
Yes None Yes Yes None Yes None None Yes None Yes Yes None None Yes Yes No Yes Unknown Yes Unknown Yes YES Yes None None Yes Yes Yes Yes Yes None Yes Yes No
Yes Yes No Yes Yes Yes Yes Yes Yes No Yes Yes No Yes Unknown Yes Unknown Yes Unknown Yes Unknown ES ES
Unknown Yes Yes ii Unknown Unknown Unknown No No
--
IPD =
Comment COPD, CHF. Subsequent concentration of 44.8 mg/L was associated with syncope. Metastatic cancer, CHF, died 2 days after theophylline concentration was obtained. (See text.) Received 2 preparations of theophylline. Received 2 preparations of theophylline plus erythromycin. Asthma exacerbation, tachycardia. Syncope, shock, COPD, intubation. Died 16 days after open heart surgery. Excessive aminophylline had been administered. COPD, CHF. COPD, CHF. Mav have innestec theoohvlline in addition to orescribed dose. Atdal fibrillason: receivid 2 preparations of thwphylline. Taking a dose of Theo-Dur that was greater than prescribed. CHF, COPD, erythromycin. COPD, CHF, CRF. CHF. COPD. COPD, ranitidine I?). COPD. COPD, arrhythmias (PVCs, PACs). Hepatitis, erythromycin. Liver disease. Suicide attemd: samole obtained 18 hours after ineestion. . Suicide gesture. COPD. PVCs. PACs. Unkndwn dose. Sinus tachycardia, anxiety. ER patient. Tachycardia; nursing home transfer. Medication confusion suspected. Syncope, sinus tachycardia. CHF; died 10 days later. Cimetidine. Agitation; deliberate self-ingestion of Theo-Dur. COPD, CHF! died 1 day later. COPD, cardrac arrest, died. Sinus tachycardia. Patient exceeded prescribed dosage of 300 mg/bid. Anxious, acute asthma. 8 hours after last dose. Grand mal seizure. Unauthorized ingestion of Theo-Dur suspected. Probable prior ingestion of theophylline.
chmnlc obstructive pulmonary disease; CHF = congestive heart tallure; CRF = chronic resplratoty failure; = emergency room.
PVCs =
premature ventricular contractions;
PACs =
premature atrial contractlons;
ophylline toxicity (concentration27.9mg/L) could not be excluded.Patient 1,who had multiple hospital admissionsfor exacerbationsof congestiveheart failure and chronic obstructive pulmonary disease, presentedwith chest pain, elevatedarterial carbon dioxide tension, and a serum theophylline concentration of 31.8 mg/L. In a subsequenthospitalization for syncope,the patient had an admissiontheophylIine concentrationof 44.8 mg/L. After review of the patient records, apparently no additional causefor the syncopewas determined.
contribute to the patient’s death. Patient 7 was a 73-year-oldwomanweighing91kg who had corticosteroid-dependentasthma,coronaryartery disease, and severeaortic stenosis.After her initially elevated theophylline concentrationof 29.8mg/L wasobtained during an admissionfor asthma, the outpatient theophylline dosewas loweredto an amount that was appropriate for her clearance(Theo-Dur 200mg twice daily). She was readmitted 2 months later for aortic valve replacementand coronaryartery bypassgraft surgery. Twelve days after surgery, she developedincreasedwheezingfor which fatalities aminophylline, which had beenwithheld for several There were five fatalities among the total of 55 days,wasreinstituted at 35mg/hour after a loading patients. In two cases,theophylline appearedto doseof 250mg.A serumtheophylline concentration
July 1991 The American Journal of Medicine
Volume 91
71
ELEVATED SERUM THEOPHYLLINE CONCENTRATIONS / GREENBERGERET AL
of 26.9 mg/L was obtained 4 days later when the aminophylline infusion rate had beenreducedto 20 mg/hour just 5 hours before.The patient died that day, after developing atrial arrhythmias with a heart rate of 132/minute. The other fatality that occurred was in Patient 36, a 74-year-oldwoman whosemedical problems included laryngeal carcinoma and chronic obstructivelung disease.The administered aminophylline doseof 30 mg/hour was predicted to give a theophylline concentration of 18.5mg/L at steady state. However,shedeveloped ventricular tachycardia (rate 180 to 2OO/minute), hypotension,and a fatal cardiacarrest after 3 days (theophylline concentration29.9mg/L). An earlier theophylline concentrationof 2 mg/L on admission should have indicated likely noncompliance with medications rather than the need for more theophylline. Effects of Prospective Intervention
In three of the patients, this investigation was instrumental in stoppingadministration of two theophylline preparations or in altering physician orders. In the remaining 39 patients, either the drug was discontinued or appropriate dosage adjustments were made by the managing physician. Clearly, in Patient 1, repeatedelevatedtheophylline concentrations occurred despite communication with the attending physician. In our opinion, the decreasedtheophylline clearanceand advanced chronic obstructive lung diseasewerereasonsto decrease the theophylline dosage substantially or totally. COMMENTS To our knowledge,this investigation is the first prospectivestudy of serumtheophylline concentrations using basic pharmacokinetic principles to assesselevatedconcentrations.Clearly, diseasestates and other host factors affect theophylline elimination [14]. The narrow therapeutic range for theophylline makesit essentialthat dosesprescribedbe appropriate in individual patients if elevatedconcentrations and the associatedrisks of toxicity are to be avoided. When theophylline concentrations exceed25mg/L, seriousadverseeffectssuchasatrial and ventricular arrhythmias and seizuresare more likely to be observed.Zwillich and colleagues [l5] reported that major motor seizureswere observedin eight of 14 patients whoseplasma theophylline concentrationsexceeded25 mg/L, four of whom died. In the current study, the incidenceof theophylline concentrationsexceeding25mg/L was low, occurring in 1.08%of all concentrationsobtained overa 2-yearperiod during which time 6,368 assayswere performed. 72
July 1991 The American Journal of Medicine
Volume 91
In 23 of 33 (69.7%)fully evaluable cases,theophylline concentrationswerebelievedto havebeen predictable in that basic pharmacokinetic calculations revealedthat serum theophylline concentrations would exceed20 mg/L. In most casesof predictable elevated theophylline concentrations,the principal source of error was that the prescribed dosesof theophylline wereinappropriately high for the theophylline clearanceobservedin a given patient. In somecases,evenafter the elevatedconcentration was obtained, ineffective reductions of the theophylline dosewere instituted by the managing physician. Failure to recognizean abnormal theophylline clearanceeither due to concurrent congestiveheart failure or chronic obstructivelung disease or to the concurrent administration of erythromycin was consideredphysician error. Failure to recognizea changingclinical statusthat may decreasetheophylline elimination was also identified as an important sourceof error in prescribing appropriate theophylline doses.Pharmacy and/or nursing oversightin addition to physician error occurred in the three patients in whom two forms of theophylline were administered simultaneously. This prospective study identified a relatively high proportion of elderly patients, with 25 of 42 patients being older than 60 years.Also, significant morbidity and mortality were observedamongthe patients who developedelevatedtheophylline concentrations. The potential adverseeffects of theophylline werecombinedwith comorbid conditions suchaschronic obstructive pulmonary diseaseand/ or congestiveheart failure. While most elevated theophylline concentrationswerepredictable using a basic calculation (C, = dosagerate/predicted clearance),the estimated apparent clearancesin eight patients were evenlower than expected,especially in patients over 70 yearswith lung or cardiac disease.This information suggeststhat extra caution should be usedwhen theophylline is administered to patients more than 70 years of age. Theophylline is well tolerated by many patients and is usually quite safe in terms of long-term administration. This current prospective evaluation revealedthat just 1.08%of theophylline determinations wereat a concentrationlikely to be associated with seriousadversereactions.Adversereactionsto theophylline canbelife-threatening or fatal despite their relative infrequency [15,16].On the basis of this study, wemake the following recommendations with respectto appropriate theophylline prescribing: (1) Patients receiving theophylline preparations should be prescribed dosesthat are determined based on considerations of theophylline clearance.Ideally, the physician should estimate the dosagerate (mg/hour) of theophylline as pre-
ELEVATED
dieted clearance (L/kg/hour) times desired steady-state plasma concentration (mg/L). (2) Optimal theophylline concentrations should be within the therapeutic range for the drug, such as 8 to 15 mg/L, to achieve efficacy and avoid toxicity, especially in patients greater than 60 years of age who have concomitant medical conditions. (3) Clearances should be assumed to be decreased in patients who are 70 years and older and/or in those who have evidence of congestive heart failure, chronic obstructive pulmonary disease, liver disease, and so forth. (4) Special attention should be paid to a patient’s changing clinical status, with regard to drug prescribing. Reductions of theophylline dosage will be indicated to avoid elevated theophylline concentrations. (5) Nursing and pharmacy personnel should also be trained in the proper use of theophylline and its salts to help avoid toxic accumulations of the drug. We believe that awareness of and vigilance in searching for the aforementioned factors associated with a decreased theophylline clearance, as well as the performance of simple pharmacokinetic calculations, will help improve theophylline prescribing. Suggested nomograms have been applied successfully to determine appropriate theophylline dosage rates [17]. Such considerations should help reduce the incidence of elevated theophylline concentrations and the associated risk of theophylline toxicity even lower than the 1.08% incidence identified. A few elevated theophylline concentrations occurred in the setting of combined oversight by physicians, nurses, and pharmacists. Such unexpected occurrences emphasize the need for continued awareness
SERUM
THEOPHYLLINE
CONCENTRATIONS
/ GREENBERGER
ET AL
by all medical personnel of even the most fundamental issues regarding theophylline.
REFERENCES 1. Mitenko PA, Ogilvie RI. Rational intravenous doses of theophylline. N Engl J Med 1973; 289: 6@&3. 2. Piafsky KM, Ogilvie RI. Dosage of theophylline in bronchial asthma. N Engl J Med 1975; 292: 1218-22. 3. Hendeles L. Massanari M, Weinberger M. Theophylline. In: Evans WE, Schentag JJ, Jusko WJ, editors. Applied pharmacokinetics: principles of therapeutic drug monitoring Spokane, Washington: Applied Therapeutics Inc., 1986: 1105-88. 4. Hendeles L. Massanari M. Weinberger M. Theophylline. In: Middleton E Jr, Reed CE. Ellis EF, Adkinson NF, Yunginger JW. editors. Allergy, principles and practice. St. Louis: CV Mosby, 1988: 673-714. 5. Gibaldi M, Perrier
D. Pharmacokinetics.
New York: Marcel
Dekker,
1982:
l-43. 6. Cohen RM. A pharmacokinetic approach to the use of theophylline in status asthmaticus. Ann Allergy 1985: 54: l-8. 7. Hendeles L, Weinberger M. Theophylline: a “state of the art” review. Pharmacotherapy 1983; 3: 2-44. 8. Harless GR. Clinical use of serum theophylline assays in a community hospital. Am J Hosp Pharm 1981; 38: 901-2. 9. Guernsey BG. lngrim NB. Hokanson JA. eta/. A utilization review of theophylline assays: sampling patterns and use. Drug Intel1 Clin Pharm 1984; 18: 906-12. 10. Sargenti C, Zelman L. Beauclair T, Garrard E. Boehm R. Evaluation of appropriateness and interpretation of serum theophylline assays. Drug Intel1 Clin Pharm 1985; 19: 380-4.
11. Gentry SM. Keith TD. McMillan DM. Goetter WE. Evaluation of the ordering of serum theophylline concentrations. Am J Hosp Pharm 1981: 38: 1937-9. 12. Winter ME, Herfindal ET, Bernstein LR. Impact of decentralized pharmacokinetics consultation service. Am J Hosp Pharm 1986; 43: 2178-84. 13. Antal EJ, Kramer PA, Mercik SA. Chapron DJ. Lawson IR. Theophyllme pharmacokinetics in advanced age. Br J Clin Pharmacol 1981; 12: 637-45. 14. Jenne JW. Effect of disease states on theophylline elimination. J Allergy Clin lmmunol 1986; 78: 727-35. 15. Zwillich CW, Sutton FD. Neff TA. Cohn WM. Matthay RA. Weinberger MM. Theophylline-induced seizures in adults. Correlation with serum concentrations. Ann Intern Med 1975; 82: 784-7. 16. Jacobs MH, Senior RM. Kessler G. Clinical experience with theophylline. JAMA 1976; 235: 1983-6. 17. Jusko WJ, Koup JR, Vance JW, Schentag JJ, Kuritzky P. Intravenous theophylline therapy: nomogram guidelines. Ann Intern Med 1977; 86: 400-4.
July
1991
The American
Journal
of Medicine
Volume
91
73
PURPOSE: To evaluate
prospectively whether serum theophylline concentrations of 25 mg/L and greater were predictable (and presumably preventable) by use of basic pharmacohinetic WlCulatiOM.
DEWGN: Prospective PATENTS Fifty-five
study. patients with a serum theophylline concentration of at least 25.0 n&L were evaluated initially and if subsequent elevated theophylline concentrations occurr’ed. INTERVENTIONS: The predicted steady-state serum theophylline concentration was calculated from the dosage rate divided by the predicted clearance to determine how many elevated concentrations (greater than 20 mg/L) were predictable. Predicted clearances were 0.04 L/l&hour for normal subjects less than 70 years of age and 0.02 L/hg/hour for patients with congestive heart failure, chronic obstructive pulmonary disease, or liver disease. Esthnated clearances were determined and compared with predicted clearances. If patients did not have steady-state concentrations, additional calculations were made. MAIN RESULTS: Prom 6968 consecutive theophylliue determinations, 69 (1.08 % ) samples from 55 patients were 25 mg/L or higher. Predictably high concentrations occurred in 23 of 33 (69.7%) fully evaluable cases. These concentrations occurred because of a failure to consider decreased elimination clearance from congestive heart failure, chronic obstructive pulmonary disease, or hepatic disease. Five fatalities occurred, and iu two cases, theophylline appeared to contribute to the patient’speak Three other patients experienced syncope. The predicted elimination clearance of theophylline of 0.02 From the Section of Allergy-Immunology, Department of Medicine, Northwestern University Medical School, Chicago, Illinois. This work was supported by United States Public Health Service Grant Al 11403 and the Ernest S. Bazley Grant. Requests for reprints should be addressed to Paul A. Greenberger, M.D., Northwestern University Medical School, 303 East Chicago Avenue, Chicago, Illinois 60611-3010. Manuscript submitted January 4.1991, and accepted March 28,199l.
L/kg/hour was too high in eight patients over 70 years old with cardiac or pulmonary disease. Nursing and pharmacy oversights were identified as three patients were given two theophylline products simultaneously. CONCLUSIONS: Most elevated theophylline concentrations are predictable (and preventable) by basic pharmacohinetic calculations. Patients experiencing elevated theophylline concentrations often had comorbid conditions and were greater thau 60 years of age. The dosage rate of theophylline (n&hour) can be estimated from predicted clearance (L/l&hour) times desired steady-state serum concentration (n&L).
T
heophylline is a bronchodilator that has been recognized to be effective in the therapy of acute and especially chronic asthma and is known to have respiratory stimulatory properties as well. Although the therapeutic response to theophylline increases proportionately with the log of the serum concentration, the risk of toxicity also increases with higher concentrations, creating a relatively narrow therapeutic range for the drug [1,2]. Potential adverse effects from theophylline are well recognized and range from minor central nervous system stimulation and gastrointestinal symptoms such as anorexia, abdominal discomfort, nausea, and vomiting to ventricular arrhythmias and seizures [2]. As the greatest therapeutic response is achieved with serum theophylline concentrations greater than 10 mg/L, and serious adverse effects are most likely to be observed at concentrations greater than 20 mg/L, 10 to 20 mg/L is the generally accepted optimal therapeutic range for the drug
WI.
Serum theophylline concentrations may be estimated by knowing the patient’s weight and theophylline dosage, and by awareness of concomitant conditions or medications that are known to affect theophylline clearance [5-71. Inappropriately elevated concentrations are undesirable and pose a risk of theophylline toxicity to the patient, although some patients are asymptomatic despite substanJuly 1991 The American Journal of Medicine
Volume 91
67
tial elevations of serum theophylline concentrations. Elevated serum theophylline concentrations can occur for many reasons, which may include patient, nursing, pharmacy, and/or physician errors [8-121, which suggests that they may be preventable. We undertook a prospective evaluation of elevated serum theophylline concentrations in an attempt to determine the incidence and causes of elevated concentrations at Northwestern Memorial Hospital, a large teaching hospital. Specific attention was directed at whether serum theophylline concentrations greater than 25 mg/L were predictable (and presumably preventable) or nonpredictable using basic pharmacokinetic calculations.
PATIENTS AND METHODS We prospectively identified all patients who had serum theophylline concentrations exceeding 25 mg/L reported at the Northwestern Memorial Hospital Clinical Pharmacology Laboratory between September 1,1988, and August 30,199O. All assays were reported by the Clinical Pharmacology Laboratory and were performed using a fluorescence polarization immunoassay with an interassay coefficient of variation of less than 5%. The reported concentrations were obtained from patients from all areas of the hospital, including inpatient wards, outpatient clinics, and the emergency room. Cases were evaluated by review of the patient records, and in some cases by additionally interviewing the patient, on the same day that the elevated concentration was reported or shortly thereafter. The investigation was not announced to medical, nursing, or pharmacy staff but was known only to investigators and members of the Pharmacy and Therapeutics Committee, Northwestern Memorial Hospital, Chicago, Illinois. Salient features that were reviewed and recorded for each case included (1) the time and date that the elevated concentration was obtained as well as its temporal relation to the last dose of theophylline, (2) the theophylline formulation and dose being taken, (3) the length of time that the patient had been taking that dose, (4) concurrent medications, (5) patient characteristics including age, height, weight, smoking status, current and previous medical problems or illnesses, reason for current hospitalization, or current clinical status, (6) the presence or absence of signs or symptoms of theophylline toxicity, and (7) previous serum theophylline concentrations and dosages. Basic pharmacokinetic analysis was performed in each patient, including (1) prediction of theophylline clearance, (2) estimation of the actual clearance in each patient, and (3) prediction of steady-state
68
July 1991
The American Journal of Medicine
Volume 91
serum theophylline tered dose.
concentration
at the adminis-
Prediction of Theophylline Clearance The clearance was predicted by knowing the patient’s weight and using previously reported average theophylline clearances that are associated with certain patient characteristics. For patients with congestive heart failure, chronic obstructive pulmonary disease, liver disease, viral upper respiratory illness, or the use of agents such as cimetidine, erythromycin, and allopurinol, an average clearance of 0.02 L/kg/hour was assumed [3]. For healthy, nonsmoking adult patients less than 70 years of age, an average clearance of 0.04 L/kg/hour was assumed. For otherwise healthy, nonsmoking adults of 70 years and older, a predicted clearance of 0.02 L/kg/hour was used [13]. For patients in whom factors that are known to be associated with increased theophylline clearance are present, such as smoking together with age less than 40 years or the use of drugs such as phenobarbital, phenytoin, or rifampin, an average clearance of 0.07 L/kg/hour was assumed [3]. Estimation of Patient Theophylline Clearance and Prediction of Serum Concentrations The patients’ estimated theophylline clearances were determined using the formula: elimination clearance (Cl) = theophylline dosage rate/serum concentration (C,,) at steady state [5]. These concentrations were considered to be at steady state with regard to theophylline pharmacokinetics (1) if medical records revealed no interruption of maintenance theophylline therapy or (2) if theophylline was initiated or the dosage was changed, and the drug had been administered for a minimum of four half-lives based on the predicted clearance for each patient. We used the formula ti/z = 0.693 V&l, where Vn is the volume of distribution calculated from 0.45 L/kg. In non-steady-state cases, the concentration of theophylline at time in hours was estimated from the following equation: C, =
loading dose v eekt D +
infusio;
rate (1 _ e-kt)
When sequential sera were obtained after the infusion was stopped, the elimination rate constant k was determined when we estimated t1/2. When additional sera were not obtained or were not helpful, we calculated k = Cl/Vn using the predicted clearance.
ELEVATED
After comparing the estimated theophylline clearancewith the initial prediction basedon literature values, we estimated the steady-stateserum theophylline concentration by the formula C,, = theophylline dosage rate (mg/hour)/estimated clearance(L/hour). When steady-stateconditions were not present,we estimated appropriatenessof dosagesand estimated serum theophylline concentrations such as with the formula: A concentration = A theophylline dose/volumeof distribution if a loading dosehad been given. If the drug had been discontinued,after apparent steadystate had been reached,we attempted to determine the half-life of theophylline if subsequentserum sampleswereobtained. Suchestimateswerecomparedwith a calculated Ti/2 of (0.693X Vn)/Cl. We consideredthe reported elevatedconcentration predictable (and thus potentially preventable) when basic pharmacokinetic calculations resulted in expected theophylline concentrations greater than 20 mg/L. Unfortunately, of 55 patientswho had serumtheophylline concentrationsof 25 mg/L or greater,13 patients are not included in the pharmacokinetic analysis because necessary clinical information (weight, theophylline preparation, dosage,and so forth) was unavailable and unretrievable. The mean theophylline concentration in those 13 patients was 28.9f 4.2 mg/L (n = 14).Five of 14sera were from the emergency room or ambulatory clinics. RESULTS Of 6,368theophylline concentrationsreported at NorthwesternMemorial Hospital betweenSeptember 1,1988,and August 30,1990,69determinations (1.08%)were greater than 25 mg/L in 55 patients. The mean of the 69 elevated serum theophylline concentrationswas 29.0 f 3.9 mg/L. Five patients (9.1%)died; in two cases(Patients 7 and 35), theophylline toxicity probably contributed to the patient’s death. Three other patients (5.5%)experienced syncopal episodes.One patient presented with a grand mal seizure.Just two elevatedconcentrations were from a recognizedsuicidal gestureor attempt. Pertinent features of eachcaseare listed in Tables I and II, including patient age,predicted and measuredserum theophylline concentrations, theophylline preparation, dosebeing taken at the time the elevated concentration was discovered, predicted and estimated theophylline clearance data, and whether there wasevidenceof toxicity to the patient from theophylline. The elevatedserum theophylline concentrationswere believedto have beenpredictable in 23 (69.7%)of the 33 patients. In
SERUM
THEOPHYLLINE
CONCENTRATIONS
/ GREENBERGER
ET AL
10 patients in whom adequatedata were obtained, our estimatesdid not predict a serum theophylline concentration of more than 20.0 mg/L. We were unable to predict in nine other casesbecauseof partial data such as absenceof weight or dosages, which limited assessment. Predictable Elevations of Theophylline
Theophylline concentrationswere elevated predictably in 23 patients. The major explanation for such elevated concentrations was that the prescribeddoseof theophylline wasexcessivein view of decreasedtheophylline clearance.In most cases, the clearancewasbelievedto be abnormal because of the presenceof congestiveheart failure and/or chronic obstructive pulmonary disease.Patient 1 had an elevatedtheophylline concentrationthat superficially would not have been predictable based on theophylline clearancesthat were calculated from concentrationspreviously obtained from this patient. However,at the time of the initial elevated concentration of 31.8mg/L, the patient had been admitted for an exacerbationof congestiveheart failure and chronic obstructive pulmonary disease. His rapidly deterioratingclinical condition had not led to an associatedreduction of the theophylline dosage.Sevenelevatedtheophylline concentrations wereobtainedovera period of 2 months subsequent to the initially elevated concentration during admissions for recurrent exacerbationsof the aforementioned medical problems.Each of thesesubsequent concentrationswaselevatedpredictably. For Patient 1, the highest serum theophylline concentration was44.8mg/L and clinically wasassociated with a syncopal episode. Patients 3,4, and 11wereinadvertently administered two forms of theophylline. Four patients (Patients 12, 30, 34, and 37) ingested theophylline in greaterthan authorized doses. Nonpredictable
Elevations of Theophylline
Elevated concentrationsobtained in 10 patients werebelievednot to havebeenpredictablebasedon predicted theophylline clearances.In nine additional patients, incompletedata made a meaningful estimate impossible. The high serum theophylline concentrationsin Patients 22 and 24 occurredas a result of suicidegesturesor attempts. Patient 6 had a history of chronic obstructive pulmonary disease and cardiacarrhythmias, and was admitted after a syncopalepisodewith shockand respiratoryfailure. A reducedtheophylline clearancefrom her compromised cardiac and pulmonary status likely explained the elevated serum concentration of 26.9 mg/L and estimated clearanceof 0.013L/kg/hour.
July
1991
The American
Journal
of Medicine
Volume
91
69
ELEVATED
SERUM
THEOPHYLLINE
CONCENTRATIONS
/ GREENBERGER
ET AL
TABLE I !Mxted Pharmacokinetic Parametersof 42 Patients with Serum Theophylline ConcentrationsGreater Than 25 mg/L Measured Theophylline Concentration*
Predicted Theophylline Concentration
Predicted Theophylline Clearance
Estimated Theophylline Clearance
(mg/Ll
(mgll)
Wk@r)
kwllr)
Theo-Dur 600 mg tid
31.8
39.2
0.020
0.024
Theo-Dur 300 Aminophylline Theo-Dur 300 Choledyl 200
25.6 32.5
20.4 31.3
0.040 0.020
0.028 0.019
27.5
23.8
0.020
0.017
30.6 26.9 29.8
16.7 17.2 28.8
0.040 0.020 0.020
0.022 0.013 0.019
mg tid mg tid 40 drip mg/hr mg x 1
30.6 25.8 25.0 27.9
35.7 24.4 21.6 17.8
0.020 0.020 0.040
0.022 0.019 0.028
Theo-Dur 300 mg qid Aminophyllinedrip 30 mg/hr
26.0 25.5
22.2 25.0
0.020 0.040
0.017 0.039
Theo-Dur 300 mg bid
26.3 25.0
21.5 28.4
0.020
0.015 0.022
Theo-Dur 300 mg tid bid
25.8 33.7
la.2 38.0
0.020
0.014 0.022
Theo-Dur 300 mg bid
25.7 32.0
20.5 32.0
0.020
0.016 0.020
Aminophylline 35 290mg/hr mg loading dose, 29 mg/hr infusion
25.3 25.7
20.3 19.6
0.020
0.016 0.015
Theo-Dur Unknown 300 number mg ofbidtablets
28.9 30.4
Unknown 21.0 dose
0.020 0.040
0.015
Unknown numberof 60 mg/hr tablets
28.9 28.1
Unknown 26.7 dose
0.020 0.040
0.020
&Q;;;ylline 500 mg tid Theo-Dur
44.9 39.4
Unknown 54.8 dose
0.020
0.023
1
Theo-Dur Unknown 200 mg bid
31.0 25.2
Unknown 14.4 dose
0.020
060
Product Dosage
mg bid drip 25 mg/hr mg bid mg bid
Theo-Dur 200 300 mg tid bid Theo-Dur 300 mg tid Theo-Dur 200 250 Theo-Dur 300 Aminophylline Theo-Dur 300
ZJT
Theo-Dur 300 200 mg tid bid
27.9 32.3
46.8 15.7
0.020
oiio
;i ii;
Aminophylline Theo-Dur 300 35 bid mg mg/hr Am;eph;;F 30 mglhr surreptitious
31.3 25.8 31.4
25.7 40.5
0.020 0.020
0.016 -
2
7:
Aminophylline 25 30 mglhr mg/hr
29.9 37.6
20.0 la.5
0.020
0.012 0.013
z; iz
2 17
0.040
0.040
0.026 -
59 :z
29.1 26.8 28.6 27.8 25.9 32.6
Unknown 11.0 Unknown Unknown
i:
Theo-Dur dose/unknown 800 mg/day Theo-Dur 300 mg tid Theo-Dur dose/unknown Theo-Dur 200 30 mg bid Aminophylline mglhr
1;::
0.020 0.040
-
0.037
Mean 28.4 f 6.5 tid = three times per day; bid = hvo times per day; qid = four times per day. *The initial theophyilineconcentration greater than 25 mgllfor each patient is presented.
Our predictedclearancesof either 0.04L/kg/hour or 0.02L/kg/hour exceededestimated clearances(Table I) for eight additional patients (Patients5,10,17, 20, 29, 31, 36, and 41). Patient 41 was apparently indiscreet in terms of theophylline use. Her usual dose was Theo-Dur@200 mg twice daily, which would result in a steady-stateserum concentration of 7.8 mg/L. The concentrationon admissionwas 25.9mg/L. Treatment with theophyllinewasdiscontinued andthen later resumed.At 200mg twice daily when steadystate occurred,her serumtheophylline concentrationwas 8.3 mg/L. We believethis caseis
70
July
1991 The American
Journal of Medicine
Volume 91
consistent with unauthorized self-overmedication with theophyllinein a patient with increasedwheezing during an asthma exacerbation. Manifestations
of Theophylline
Toxicity
Evidence of theophylline toxicity as noted in 25 (59.5%)of the 42 patients (Table II) consistedof nausea,jitterness, palpitations, tachycardia, arrhythmias, grandmal seizure,and syncope.Patient 11,who had a history of atrial fibrillation, presented with atrial fibrillation and rapid ventricular response.An associationof atrial fibrillation with the-
ELEVATED
SERUM
THEOPHYLLINE
CONCENTRATIONS
/ GREENBERGER
ET AL
TABLE II Initial Theophylline Concentrationsin 42 Patients and Potential Theophylline Toxicity
Patient Number
-
Age
Measured Theophylline Concentration (mglL)
Potential Toxkieb;t ldentiiication
Predictable Concentrations Greaterthan 20 m%L
31.8
None
Yes
32.5
Yes
Yes
25.6 27.5 30.6 26.9 29.8
Yes None Yes Possible Yes
Yes Yes No Unknown Yes
30.6 25.8 25.0 27.9 25.5 26.0 25.0 26.3 33.7 25.8 32.0 25.7 25.7 25.3 30.4 28.9 28.1 28.9 39.4 44.9 25.2 31.0 27.9 32.3 31.3 25.8 31.4 37.6 29.9 29.1 26.8 28.6 27.8 25.9 32.6
Yes None Yes Yes None Yes None None Yes None Yes Yes None None Yes Yes No Yes Unknown Yes Unknown Yes YES Yes None None Yes Yes Yes Yes Yes None Yes Yes No
Yes Yes No Yes Yes Yes Yes Yes Yes No Yes Yes No Yes Unknown Yes Unknown Yes Unknown Yes Unknown ES ES
Unknown Yes Yes ii Unknown Unknown Unknown No No
--
IPD =
Comment COPD, CHF. Subsequent concentration of 44.8 mg/L was associated with syncope. Metastatic cancer, CHF, died 2 days after theophylline concentration was obtained. (See text.) Received 2 preparations of theophylline. Received 2 preparations of theophylline plus erythromycin. Asthma exacerbation, tachycardia. Syncope, shock, COPD, intubation. Died 16 days after open heart surgery. Excessive aminophylline had been administered. COPD, CHF. COPD, CHF. Mav have innestec theoohvlline in addition to orescribed dose. Atdal fibrillason: receivid 2 preparations of thwphylline. Taking a dose of Theo-Dur that was greater than prescribed. CHF, COPD, erythromycin. COPD, CHF, CRF. CHF. COPD. COPD, ranitidine I?). COPD. COPD, arrhythmias (PVCs, PACs). Hepatitis, erythromycin. Liver disease. Suicide attemd: samole obtained 18 hours after ineestion. . Suicide gesture. COPD. PVCs. PACs. Unkndwn dose. Sinus tachycardia, anxiety. ER patient. Tachycardia; nursing home transfer. Medication confusion suspected. Syncope, sinus tachycardia. CHF; died 10 days later. Cimetidine. Agitation; deliberate self-ingestion of Theo-Dur. COPD, CHF! died 1 day later. COPD, cardrac arrest, died. Sinus tachycardia. Patient exceeded prescribed dosage of 300 mg/bid. Anxious, acute asthma. 8 hours after last dose. Grand mal seizure. Unauthorized ingestion of Theo-Dur suspected. Probable prior ingestion of theophylline.
chmnlc obstructive pulmonary disease; CHF = congestive heart tallure; CRF = chronic resplratoty failure; = emergency room.
PVCs =
premature ventricular contractions;
PACs =
premature atrial contractlons;
ophylline toxicity (concentration27.9mg/L) could not be excluded.Patient 1,who had multiple hospital admissionsfor exacerbationsof congestiveheart failure and chronic obstructive pulmonary disease, presentedwith chest pain, elevatedarterial carbon dioxide tension, and a serum theophylline concentration of 31.8 mg/L. In a subsequenthospitalization for syncope,the patient had an admissiontheophylIine concentrationof 44.8 mg/L. After review of the patient records, apparently no additional causefor the syncopewas determined.
contribute to the patient’s death. Patient 7 was a 73-year-oldwomanweighing91kg who had corticosteroid-dependentasthma,coronaryartery disease, and severeaortic stenosis.After her initially elevated theophylline concentrationof 29.8mg/L wasobtained during an admissionfor asthma, the outpatient theophylline dosewas loweredto an amount that was appropriate for her clearance(Theo-Dur 200mg twice daily). She was readmitted 2 months later for aortic valve replacementand coronaryartery bypassgraft surgery. Twelve days after surgery, she developedincreasedwheezingfor which fatalities aminophylline, which had beenwithheld for several There were five fatalities among the total of 55 days,wasreinstituted at 35mg/hour after a loading patients. In two cases,theophylline appearedto doseof 250mg.A serumtheophylline concentration
July 1991 The American Journal of Medicine
Volume 91
71
ELEVATED SERUM THEOPHYLLINE CONCENTRATIONS / GREENBERGERET AL
of 26.9 mg/L was obtained 4 days later when the aminophylline infusion rate had beenreducedto 20 mg/hour just 5 hours before.The patient died that day, after developing atrial arrhythmias with a heart rate of 132/minute. The other fatality that occurred was in Patient 36, a 74-year-oldwoman whosemedical problems included laryngeal carcinoma and chronic obstructivelung disease.The administered aminophylline doseof 30 mg/hour was predicted to give a theophylline concentration of 18.5mg/L at steady state. However,shedeveloped ventricular tachycardia (rate 180 to 2OO/minute), hypotension,and a fatal cardiacarrest after 3 days (theophylline concentration29.9mg/L). An earlier theophylline concentrationof 2 mg/L on admission should have indicated likely noncompliance with medications rather than the need for more theophylline. Effects of Prospective Intervention
In three of the patients, this investigation was instrumental in stoppingadministration of two theophylline preparations or in altering physician orders. In the remaining 39 patients, either the drug was discontinued or appropriate dosage adjustments were made by the managing physician. Clearly, in Patient 1, repeatedelevatedtheophylline concentrations occurred despite communication with the attending physician. In our opinion, the decreasedtheophylline clearanceand advanced chronic obstructive lung diseasewerereasonsto decrease the theophylline dosage substantially or totally. COMMENTS To our knowledge,this investigation is the first prospectivestudy of serumtheophylline concentrations using basic pharmacokinetic principles to assesselevatedconcentrations.Clearly, diseasestates and other host factors affect theophylline elimination [14]. The narrow therapeutic range for theophylline makesit essentialthat dosesprescribedbe appropriate in individual patients if elevatedconcentrations and the associatedrisks of toxicity are to be avoided. When theophylline concentrations exceed25mg/L, seriousadverseeffectssuchasatrial and ventricular arrhythmias and seizuresare more likely to be observed.Zwillich and colleagues [l5] reported that major motor seizureswere observedin eight of 14 patients whoseplasma theophylline concentrationsexceeded25 mg/L, four of whom died. In the current study, the incidenceof theophylline concentrationsexceeding25mg/L was low, occurring in 1.08%of all concentrationsobtained overa 2-yearperiod during which time 6,368 assayswere performed. 72
July 1991 The American Journal of Medicine
Volume 91
In 23 of 33 (69.7%)fully evaluable cases,theophylline concentrationswerebelievedto havebeen predictable in that basic pharmacokinetic calculations revealedthat serum theophylline concentrations would exceed20 mg/L. In most casesof predictable elevated theophylline concentrations,the principal source of error was that the prescribed dosesof theophylline wereinappropriately high for the theophylline clearanceobservedin a given patient. In somecases,evenafter the elevatedconcentration was obtained, ineffective reductions of the theophylline dosewere instituted by the managing physician. Failure to recognizean abnormal theophylline clearanceeither due to concurrent congestiveheart failure or chronic obstructivelung disease or to the concurrent administration of erythromycin was consideredphysician error. Failure to recognizea changingclinical statusthat may decreasetheophylline elimination was also identified as an important sourceof error in prescribing appropriate theophylline doses.Pharmacy and/or nursing oversightin addition to physician error occurred in the three patients in whom two forms of theophylline were administered simultaneously. This prospective study identified a relatively high proportion of elderly patients, with 25 of 42 patients being older than 60 years.Also, significant morbidity and mortality were observedamongthe patients who developedelevatedtheophylline concentrations. The potential adverseeffects of theophylline werecombinedwith comorbid conditions suchaschronic obstructive pulmonary diseaseand/ or congestiveheart failure. While most elevated theophylline concentrationswerepredictable using a basic calculation (C, = dosagerate/predicted clearance),the estimated apparent clearancesin eight patients were evenlower than expected,especially in patients over 70 yearswith lung or cardiac disease.This information suggeststhat extra caution should be usedwhen theophylline is administered to patients more than 70 years of age. Theophylline is well tolerated by many patients and is usually quite safe in terms of long-term administration. This current prospective evaluation revealedthat just 1.08%of theophylline determinations wereat a concentrationlikely to be associated with seriousadversereactions.Adversereactionsto theophylline canbelife-threatening or fatal despite their relative infrequency [15,16].On the basis of this study, wemake the following recommendations with respectto appropriate theophylline prescribing: (1) Patients receiving theophylline preparations should be prescribed dosesthat are determined based on considerations of theophylline clearance.Ideally, the physician should estimate the dosagerate (mg/hour) of theophylline as pre-
ELEVATED
dieted clearance (L/kg/hour) times desired steady-state plasma concentration (mg/L). (2) Optimal theophylline concentrations should be within the therapeutic range for the drug, such as 8 to 15 mg/L, to achieve efficacy and avoid toxicity, especially in patients greater than 60 years of age who have concomitant medical conditions. (3) Clearances should be assumed to be decreased in patients who are 70 years and older and/or in those who have evidence of congestive heart failure, chronic obstructive pulmonary disease, liver disease, and so forth. (4) Special attention should be paid to a patient’s changing clinical status, with regard to drug prescribing. Reductions of theophylline dosage will be indicated to avoid elevated theophylline concentrations. (5) Nursing and pharmacy personnel should also be trained in the proper use of theophylline and its salts to help avoid toxic accumulations of the drug. We believe that awareness of and vigilance in searching for the aforementioned factors associated with a decreased theophylline clearance, as well as the performance of simple pharmacokinetic calculations, will help improve theophylline prescribing. Suggested nomograms have been applied successfully to determine appropriate theophylline dosage rates [17]. Such considerations should help reduce the incidence of elevated theophylline concentrations and the associated risk of theophylline toxicity even lower than the 1.08% incidence identified. A few elevated theophylline concentrations occurred in the setting of combined oversight by physicians, nurses, and pharmacists. Such unexpected occurrences emphasize the need for continued awareness
SERUM
THEOPHYLLINE
CONCENTRATIONS
/ GREENBERGER
ET AL
by all medical personnel of even the most fundamental issues regarding theophylline.
REFERENCES 1. Mitenko PA, Ogilvie RI. Rational intravenous doses of theophylline. N Engl J Med 1973; 289: 6@&3. 2. Piafsky KM, Ogilvie RI. Dosage of theophylline in bronchial asthma. N Engl J Med 1975; 292: 1218-22. 3. Hendeles L. Massanari M, Weinberger M. Theophylline. In: Evans WE, Schentag JJ, Jusko WJ, editors. Applied pharmacokinetics: principles of therapeutic drug monitoring Spokane, Washington: Applied Therapeutics Inc., 1986: 1105-88. 4. Hendeles L. Massanari M. Weinberger M. Theophylline. In: Middleton E Jr, Reed CE. Ellis EF, Adkinson NF, Yunginger JW. editors. Allergy, principles and practice. St. Louis: CV Mosby, 1988: 673-714. 5. Gibaldi M, Perrier
D. Pharmacokinetics.
New York: Marcel
Dekker,
1982:
l-43. 6. Cohen RM. A pharmacokinetic approach to the use of theophylline in status asthmaticus. Ann Allergy 1985: 54: l-8. 7. Hendeles L, Weinberger M. Theophylline: a “state of the art” review. Pharmacotherapy 1983; 3: 2-44. 8. Harless GR. Clinical use of serum theophylline assays in a community hospital. Am J Hosp Pharm 1981; 38: 901-2. 9. Guernsey BG. lngrim NB. Hokanson JA. eta/. A utilization review of theophylline assays: sampling patterns and use. Drug Intel1 Clin Pharm 1984; 18: 906-12. 10. Sargenti C, Zelman L. Beauclair T, Garrard E. Boehm R. Evaluation of appropriateness and interpretation of serum theophylline assays. Drug Intel1 Clin Pharm 1985; 19: 380-4.
11. Gentry SM. Keith TD. McMillan DM. Goetter WE. Evaluation of the ordering of serum theophylline concentrations. Am J Hosp Pharm 1981: 38: 1937-9. 12. Winter ME, Herfindal ET, Bernstein LR. Impact of decentralized pharmacokinetics consultation service. Am J Hosp Pharm 1986; 43: 2178-84. 13. Antal EJ, Kramer PA, Mercik SA. Chapron DJ. Lawson IR. Theophyllme pharmacokinetics in advanced age. Br J Clin Pharmacol 1981; 12: 637-45. 14. Jenne JW. Effect of disease states on theophylline elimination. J Allergy Clin lmmunol 1986; 78: 727-35. 15. Zwillich CW, Sutton FD. Neff TA. Cohn WM. Matthay RA. Weinberger MM. Theophylline-induced seizures in adults. Correlation with serum concentrations. Ann Intern Med 1975; 82: 784-7. 16. Jacobs MH, Senior RM. Kessler G. Clinical experience with theophylline. JAMA 1976; 235: 1983-6. 17. Jusko WJ, Koup JR, Vance JW, Schentag JJ, Kuritzky P. Intravenous theophylline therapy: nomogram guidelines. Ann Intern Med 1977; 86: 400-4.
July
1991
The American
Journal
of Medicine
Volume
91
73
