Reducing Pain Experienced During Potassium Chloride Infusion in the Emergency Department - Document (2024)

Full Text:

Introduction

Hypokalemia is a common electrolyte abnormality encountered in theclinical setting. It can lead to cardiac arrhythmias, muscle weakness, andrhabdomyolysis (1-5). The causes of hypokalemia can be broadly classifiedinto decreased intake, increased entry into cells due to causes such asincrease in extracellular pH, increased P-adrenergic activity or insulinadministration, and increased losses from sweating, and the gastrointestinalor renal systems (6). In addition to the diagnosis and treatment of theunderlying cause(s) of hypokalemia, replacement of the potassium deficit isnecessary when managing these patients. This urgency of replacement therapywill depend on the severity of hypokalemia, the rate of decline in serumpotassium concentration, and associated comorbidities of the patient.Potassium can be administered as potassium chloride, potassium phosphate,potassium bicarbonate potassium gluconate or its precursors (potassiumcitrate, potassium acetate) (7, 8). Potassium chloride has been the mainstayof replacement therapy for two main reasons:

* The serum potassium concentration increases at a faster ratethan other formulation as chloride is primarily an extracellular anion,thereby promoting maintenance of the administered potassium in theextracellular fluid (9).

* Patients with hypokalemia are often chloride-depleted. Thischloride-depleted state will contribute to metabolic alkalosis by enhancingrenal bicarbonate reabsorption that leads to potassium wasting, as sodium isreabsorbed in exchange for secreted potassium rather than with chloride (10,11).

Intravenous replacement should be used in patients with severesymptomatic hypokalemia (less than 2.5 mmol/L) and in patients with lesssevere hypokalemia who are unable to take oral medications. The most commonproblem associated with potassium chloride infusion is pain. The occurrenceof pain has been reported in up to 60% of patients receiving intravenousreplacement (12), resulting in patient's refusal of the prescribedtherapy (13). While the exact mechanism of action is not known, possiblecauses include direct chemical and/or osmotic irritation of the intima andmechanical distension from volume of drug injected (14). Prevention orreduction of pain can be achieved by running potassium chloride at aninfusion rate below 20 mmol/hour (15), injecting into large vein using alarge bore cannula, adding a diluent (e.g., 0.9% normal saline) during theinfusion process, and applying an ice pack.

Previous studies have demonstrated the efficacy of lignocaine inalleviating pain during potassium chloride infusion both immediately and overan infusion time of up to 2 hours (14-16). However, the varying doses oflignocaine, small sample size, and subsequent reports of safety issues havelimited the prevalent use of lignocaine in clinical practice. Safety issueshave been attributed to the different formulations of lignocaine availableand the extra step of preparation which led to an increased likelihood ofmedical error (17). Concerns have also been raised about the potential oflignocaine to mask phlebitis. More than a decade later, we embarked on thisstudy to relook at the efficacy and safety related to the use of lignocainein reducing pain during potassium chloride infusion. We hope that ourfindings will offer clinicians the option of a fairly simple interventionthat will address the issue of pain experienced by patients during potassiumchloride infusion.

Materials and Methods

Setting

This study was conducted in an Emergency Department (ED) inSingapore that sees about 135,000 patients a year.

Design

A prospective observational study of the effect of lignocaine onpain during potassium chloride infusion was recorded in a continuous qualityimprovement database from March 2015 to May 2016. Patients were recruited byconsecutive sampling. They were included if the attending ED physicianordered intravenous potassium replacement with potassium chloride. Patientswith a history of epilepsy, cardiac arrhythmia, and sensitivity to lignocainewere excluded from the study. Baseline characteristics of age, gender,concentration of serum potassium, duration of potassium chloride infusion,and concurrent run of normal saline were recorded prior to the initiation oflignocaine.

Potassium chloride was given as a premixed solution containing 10mmol of potassium chloride in 100 mL of water for injection. Thiscommercially available preparation was routinely used for potassium chlorideinfusion as per departmental protocol. Patients who received lignocaine weregiven a 3 mL bolus of 1% solution a minute before the start of the potassiumchloride infusion. This intervention was unblinded to the patient andattending staff. Patients were put on cardiac monitoring during the period ofintravenous replacement if they had received lignocaine.

All doctors and nurses working in the ED were briefed on thelignocaine protocol. As this practice was recommended but not astandard-of-care in the department, the decision to administer lignocaine tothe patient was left to the discretion of the attending physician.

Outcome

The primary outcome was to determine the efficacy of lignocaine inreducing pain experienced during potassium chloride infusion. The qualitativeoccurrence of self-reported pain was tracked instead of a quantitative painscore during the potassium chloride infusion. This was chosen as a practicaloutcome measure, as any occurrence of pain would necessitate interventionfrom the attending staff to help the patient tolerate the infusion. Theseinterventions were dependent on the attending physician and included applyingice pack, slowing down the initial infusion rate, and stopping the infusiondue to patient's refusal. The secondary outcome was to determine theadverse events associated with intravenous lignocaine use for this purpose.Adverse events were defined as lignocaine toxicity, cardiac arrhythmia,masking of phlebitis, and medication error. Occurrence of any pain or adverseevent was recorded by the attending staff and subsequently traced from reviewof the patients' case records.

Sample size

On the basis of the previous study by Lim et al. (14), theincidence of pain in patients who did not receive lignocaine was 85.7%compared with 28.6% when lignocaine was given. For a reduction of at least50% in the occurrence of pain in patients who received lignocaine prior tothe potassium chloride infusion, a sample size of 22 with equal numbers ineach arm was required to power our study ([alpha]=0.05, [beta]=0.2,power=0.8). To increase the robustness of the study, patients were enrolledin the study until the end of the quality improvement initiative that wouldrecruit 50 patients in each arm of the study as determined a priori.

Statistical analysis

To control for selection bias in the assignment of subjects totreatment due to the baseline characteristics, stabilized inverse probabilityof treatment weights (IPTWs) were used in the analysis to estimate theefficacy of lignocaine on pain outcome. IPTWs were derived using weights onthe basis of propensity scores to create a pseudo-sample in which thedistribution of the measured baseline covariates was independent of thetreatment assignment (18). Propensity scores of the subjects were estimatedusing the multivariable logistic regression where the status of treatment wasregressed on the measured baseline covariates. Stabilized IPTWs will reducevariability due to instability in estimation that could be induced either bythose treated subjects with low propensity scores or untreated subjects withhigh propensity scores (18, 19). Absolute standardized difference wascalculated for each of the baseline covariates to assess the balance betweenboth groups and was compared before and after applying stabilized weights. Anabsolute standardized difference of less than 0.1 was used to indicatenegligible difference in the mean or prevalence of a covariate between thetreatment and control group.

Risk estimates and numbers needed to treat (NNT) were estimatedusing stabilized weights. Chi-square test was used to test for differencesbetween categorical variables and two-sample t-test was used for continuousvariables. All statistical analyses were performed using Statistical Packagefor the Social Sciences (IBM SPSS Statistics; Armonk, NY, USA) version 24.

Results

The baseline demographics and clinical characteristics of the 100patients are presented in Table 1. The average age of patients included inthe study was 64.5 ([+ or -] 19.3) years. Majority of the patients werefemales (70%). Eighty-one percent of all patients required more than 1 hourof potassium chloride infusion, and 79% of the patients required concurrentrun of normal saline due to their presenting medical conditions. The averageserum potassium concentration of patients was 2.7 ([+ or -] 0.4) mmol/L.

Prior to stabilized weights adjustment, all covariates, exceptgender, had an absolute standardized difference of more than 0.1, with amaximum of 0.55 for the mean serum potassium concentration (p=0.007) (Table2). On average, patients in the lignocaine group tend to be younger (meanage, 63.0 vs. 66.0 years) and had a higher level of serum potassiumconcentration (2.8 vs. 2.6 mmol/L). The proportion of patients who requiredpotassium chloride infusion of more than 1 hour was lower in the lignocainegroup than in the control group (74.0% vs. 86.0%). More patients in thelignocaine group had concurrent run of normal saline solution during thepotassium chloride infusion (84.0% vs. 74.0%). There was no absolutedifference in the proportion of gender between the two groups.

After stabilized weights adjustment, the resultant sample size was98, with 49 patients in each group. There were no significant differences inthe baseline covariates between the two groups. The absolute standardizeddifferences for all covariates in the weighted sample ranged from a maximumof 0.1 to a minimum of 0.03. Balance of baseline covariates between thelignocaine and the control group was achieved in the stabilized weightsadjusted sample.

The efficacy of lignocaine on pain was estimated on the basis ofthe weighted sample where the assignment of patients to lignocaine wasindependent on the patients' baseline characteristics. The proportion ofpatients experiencing pain was significantly higher in the control groupwhere no lignocaine was given prior to the potassium chloride infusion (38.8%vs. 16.3%, p=0.013) (Figure 1).

The absolute risk reduction was 22.5% and the relative risk ofpain associated with lignocaine was 0.42 (95% C.I., 0.20-0.87, p=0.013). Inother words, the risk of experiencing pain for patients who were not assignedto the lignocaine group was 2.4 times the risk of those who were givenlignocaine. The numbers needed to treat was 5 (95% C.I., 3-19), suggestingthat on average, for every five patients who were treated with lignocaineprior to the initiation of potassium chloride infusion, one will benefit fromthe treatment.

There were no adverse events associated with the lignocaineprotocol used in this study, particularly the occurrence of cardiacarrhythmias, lignocaine toxicity, and medication errors.

Discussion

Pain control and lignocaine

Several lignocaine regimens have been used in previous studies toreduce the pain during potassium chloride infusion. Lim et al. (14) used 3 mLof 1% (30 mg) lignocaine as a bolus 1 minute before starting a 20 mmolpotassium chloride diluted in 100 ml dextrose 5% infusion which wasadministered over 2 hours. Morrill et al. (15) used 10 mg of lignocaine addedto 10 mmol of potassium chloride in 50 ml of dextrose 5% which was ran over 1hour. Pucino et al. (16) used 50 mg of lignocaine added to 20 mmol ofpotassium chloride in 50 ml of dextrose 5% which was administered over 2hours. The preparation of potassium chloride used in our institution was 10mmol in 100 mL of water for injection, which was less concentrated than thesolutions used in previous studies. Therefore, the incidence of pain in ourpatients without lignocaine was 38.8%, lower than the 60% reported inliterature. To ensure that the lignocaine dose we used was effective incontrolling pain and had minimal risk for adverse event, we decided to use 3mL of 1% lignocaine for in our study. Furthermore, Lim et al. (14) has shownthat this dose of lignocaine was effective for pain control both immediatelyand throughout a 2-hour infusion period, attributable to the duration ofaction of lignocaine [t1/2, 80-108 minutes (20, 21)]. With the increased insample size, our study had shown that lignocaine was able to reduce to painassociated with potassium chloride infusion. At S$0.74 per vial and a numberneeded to treat of 5, this intervention was inexpensive, in addition to beingsafe at the study regimen.

We would want to highlight that the technique of slow intravenouspush when administering the bolus lignocaine dose was important foranesthetizing the nerves fibers that supply the veins. A rapid bolus woulddeliver the drug systematically rather than allowing lignocaine to actlocally and may render the lignocaine bolus ineffective.

Intervention needed

Application of ice pack was done more commonly in 71.4% patientsas compared to slowing down of infusion rate in 25% patients who experiencedpain. It was unlikely that the attending ED physician would agree to slowdown the infusion rate as the first measure after making the decision toinitiate intravenous replacement therapy for severe, symptomatic hypokalemia.Our study did not standardize the use of a large vein (e.g., cubital fossa)or a particular cannula size as in previous studies. By leaving this to theattending ED physicians, a pragmatic trial would demonstrate the efficacy oflignocaine in reducing pain during potassium chloride infusion.

Adverse outcomes

While there was genuine concern for potential adverse events fromlignocaine use, this likelihood was low at the prescribed dose of 30 mg. Foran average 70 kg adult, this was 0.4 mg/kg, way less than the toxic dose of 3mg/kg for lignocaine where cardiac arrhythmias and other features oflignocaine toxicity ranging from circumoral numbness to seizures may occur.To address the concern of cardiac arrhythmias, our patients were put oncardiac monitoring throughout the period of infusion during the study. Noarrhythmia was noted with this dosage of lignocaine. Therefore, the routineuse of cardiac monitoring with lignocaine in our study's prescribed doseand use was not recommended. There were also no reports of featuresattributable to lignocaine toxicity. Phlebitis could be avoided by assessingthe injection site before running the potassium chloride infusion withlignocaine.

Another reason contributing to the infrequent use of lignocaine toreduce pain from potassium chloride infusion was medical error arising froman extra step in preparation. Errors had resulted when staff used the wrongformulation due to the different concentrations and preparations oflignocaine currently available. This should not be a reason to deny patientsa simple intervention which could be carried out by trained staff adhering tostringent preparation protocols, as with many other medications being used inany hospitals.

Study limitations

The main limitation of this observational study was the lack ofrandomization of patients into receiving lignocaine prior to potassiumchloride infusion. However, we attempted to control for this by the use ofstabilized weights in the statistical analysis. Nonetheless, unidentifiedconfounders may still exist, affecting the results.

Another limitation was that absolute occurrence of pain wastracked instead of pain score. Therefore, for patients who experienced painduring the infusion, we were unable to compare the difference in pain scoresbetween the two groups. This decision was based on an ideal outcome, as thepretreatment with lignocaine should alleviate all pain associated with theinfusion, removing the need for intervention from the attending staff. Theextent of pain score reduction with lignocaine during potassium chlorideinfusion and its effectiveness would need to be evaluated in a separatestudy.

Conclusion

The addition of lignocaine to intravenous potassium chloridereplacement was useful in reducing pain during the infusion process. Thiswould improve patient acceptance of the therapy and also reduce the need foradditional intervention from the attending staff. In addition, we have proventhat this extra step in preparation could be done safely and the addition oflignocaine was not associated with any adverse events as previously thought.Therefore, we recommend the routine use of 3 ml of 1% lignocaine priorintravenous potassium chloride replacement as a new practice standard.

DOI: 10.5152/eajem.2017.53825

Peer-review: Externally peer-reviewed.

Conflict of Interest: No conflict of interest was declared by theauthors.

Financial Disclosure: The authors declared that this study hasreceived no financial support.

References

(2.) Comi G, Testa D, Cornelio F, Comola M, Canal N. Potassiumdepletion myopathy: a clinical and morphological study of six cases. MuscleNerve 1985; 8: 17-21. [CrossRef]

(3.) Shintani S, Shiigai T, Tsukagoshi H. Marked hypokalemicrhabdomyolysis with myoglobinuria due to diuretic treatment. Eur Neurol 1991;31: 3968. [CrossRef]

(4.) Dominic JA, Koch M, Guthrie GP Jr, Galla JH. Primaryaldosteronism presenting as myoglobinuric acute renal failure. Arch InternMed 1978; 138: 1433-4. [CrossRef]

(5.) Knochel JP. Neuromuscular manifestations of electrolytedisorders. Am J Med 1982; 72: 521-35. [CrossRef]

(6.) Mount DB. Causes of hypokalemia in adults. In: UpToDate, PostTW (Ed), UpToDate, Waltham, MA. (Accessed on August 5, 2016)

(7.) Cohn JN, Kowey PR, Whelton PK, Prisant LM. New guidelines forpotassium replacement in clinical practice: a contemporary review by theNational Council on Potassium in Clinical Practice. Arch Intern Med 2000;160: 2429-36. [CrossRef]

(8.) Kim GH, Han JS. Therapeutic approach to hypokalemia. Nephron2002; 92(Suppl 1): 28. [CrossRef]

(9.) Villamil MF, Deland EC, Henney RP, Maloney JV Jr. Anioneffects on cation movements during correction of potassium depletion. Am JPhysiol 1975; 229: 161 -6.

(10.) Kassirer JP, Schwartz WB. The response of normal man toselective depletion of hydrochloric acid. Factors in the genesis ofpersistent gastric alkalosis. Am J Med 1966; 40: 10-8. [CrossRef]

(11.) Kassirer JP, Berkman PM, Lawrenz DR, Schwartz WB. Thecritical role of chloride in the correction of hypokalemic alkalosis in man.Am J Med 1965; 38: 172-89. [CrossRef]

(12.) Stephen M, Loewenthal J, Wong J, Benn R. Complications ofintravenous therapy. Med J Aust 1976; 2: 557-9.

(13.) Chan R, Armstrong D. Peripheral intravenous infusion ofpotassium chloride: effect of solution composition on infusion-site pain.Canadian J Hosp Pharm 2004; 57: 27-31.

(14.) Lim ET, Khoo ST, Tweed WA. Efficacy of lignocaine inalleviating potassium chloride infusion pain. Anaesth Intensive Care 1992;20: 196-8.

(15.) Morrill GB, Katz MD. The use of lidocaine to reduce paininduced by potassium chloride infusion. J Intraven Nurs 1988; 11: 105-8.

(16.) Pucino F, Danielson BD, Carlson JD, Strommen GL, Walker PR,Beck CL, et al. Patient tolerance to intravenous potassium chloride with andwithout lidocaine. Drug Intell Clin Pharm 1988; 22: 676-9.

(17.) Grissinger M. Adding lidocaine to IV potassium infusions cancause safety problems. P T. 2008; 33: 70-5.

(18.) Peter CA. An Introduction to Propensity Score Methods forReducing the Effects of Confounding in Observational Studies. MultivariateBehav Res 2011; 46: 399-424. [CrossRef]

(19.) Xu S, Ross C, Raebel MA, Shetterly S, Blanchette C, Smith D.Use of Stabilized Inverse Propensity Scores as Weights to Directly EstimateRelative Risk and Its Confidence Intervals. Value Health 2010; 13: 273-7.[CrossRef]

(20.) Boyes RN, Scott DB, Jebson PJ, Godman MJ, Julian DG.Pharmaco*kinetics of lidocaine in man. Clin Pharmacol Ther 1971; 12: 105-16.[CrossRef]

(21.) Rowland M, Thomson PD, Guichard A, Melmon KL. Dispositionkinetics of lidocaine in normal subjects. Ann NY Acad Sci 1971; 179: 383-98.[CrossRef]

Jen Heng Pek [1], Hui Cheng Tan [1], Puneet Seth [1], Evelyn Wong[2]

[1] Sengkang Health, Singhealth, Singapore

[2] Singapore General Hospital, Singhealth, Singapore

Correspondence to: Jen Heng Pek e-mail: jenheng_@hotmail.com

Received: 27.05.2017 * Accepted: 21.06.2017 * Available OnlineDate: 09.10.2017

Table 1. Patients' baseline demographics and clinicalcharacteristicsVariable n=100Age, mean ([+ or -] SD), years 64.5 ([+ or -] 19.3)Gender (%)Male 30.0Female 70.0Serum Potassium, mean ([+ or -] SD), mmol/L 2.7 ([+ or -] 0.4)Infusion duration required (%)= 1 hour 19.0> 1 hour 81.0Concurrent Normal Saline (%)Given 79.0Not given 21.0Table 2. Characteristics of patients for the unweighted andweighted samples Unweighted Lignocaine (n=50) No Lignocaine (n=50)Gender, n (%) 15 (30.0) 15 (30.0)MaleAge, mean ([+ or -] 63.0 ([+ or -] 19.7) 66.0 ([+ or -] 19.1)SD), yearsSerum Potassium, mean 2.8 ([+ or -] 0.3) 2.6 ([+ or -] 0.4)([+ or /] SD), mmol/LInfusion duration more 38 (76.0) 43 (86.0)than 1 hour, n (%)Concurrent Normal 42 (84.0) 37 (74.0)Saline given, n (%) Unweighted P (Std Diff)Gender, n (%) --Male (0.00)Age, mean ([+ or -] 0.441 (0.15)SD), yearsSerum Potassium, mean 0.007 (0.55)([+ or /] SD), mmol/LInfusion duration more 0.202 (0.26)than 1 hour, n (%)Concurrent Normal 0.220 (0.25)Saline given, n (%) Weighted by Stabilized IPTWs Lignocaine (n=49) No Lignocaine (n=49)Gender, n (%) 15 (30.6)Male 14 (28.6)Age, mean ([+ or -] 64.6 ([+ or -] 19.5) 65.1 ([+ or -] 18.8)SD), yearsSerum Potassium, mean 2.7 ([+ or -] 0.3) 2.6 ([+ or -] 0.4)([+ or /] SD), mmol/LInfusion duration more 38 (77.6) 40 (81.6)than 1 hour, n (%)Concurrent Normal 39 (79.6) 38 (77.6)Saline given, n (%) Weighted by Stabilized IPTWs P (Std Diff)Gender, n (%) 0.825Male (0.04)Age, mean ([+ or -] 0.883 (0.03)SD), yearsSerum Potassium, mean 0.613 (0.10)([+ or /] SD), mmol/LInfusion duration more 0.616 (0.10)than 1 hour, n (%)Concurrent Normal 0.806 (0.05)Saline given, n (%)Figure 1. Proportion of patients experiencing pain during potassiumchloride infusion no pain painWith lignocaine [n=49] 83.7% 16.3%Without lignocaine [n=49] 61.2% 38.8%Note: Table made from bar graph.