bacterial contamination was mainly found to be from Gram-negative organisms. Contamination with Gram-positive bacteria such as MRSA and VRE that frequently cause Nosocomial infections were only found on 2 keyboards. The remaining keyboards harbored mostly gram-negative organ- isms that are not well known. E coli and PSEUDO were found at low prevalence on 2 different keyboards and also commonly cause nosocomial infections. The remaining gram-negative contamination resulted from mostly PANT and 1 incidence of SPHING. Both organisms are rare but have been known to cause some nosocomial infections. It is unclear why more gram-negative contamination occurred compared with gram positive.

Conclusions

Keyboards are commonly colonized with bacteria; however, in this study, we found the prevalence of colonization with pathogenic bacteria to be low (13.8%) as compared with other studies (26%-95%). More keyboards harbored Gram-negative bacteria vs gram-positive bacteria (12.5% vs 4.2%), and no CDIF was identified. MRSA was only found on 1 keyboard, also representing a low contamination rate. Keyboards located in nontreatment areas of the ED had higher levels of bacterial contamination, possibly related to lower hand washing frequency.

Angela Pugliese MD

A. Joseph Garcia MD William Dobson

Department of Emergency Medicine

Henry Ford Hospital CFP 2, DEM, Detroit, MI 48202, USA

E-mail address: apuglie1@hfhs.org

Linoj Samuel PhD Department of Pathology Henry Ford Hospital

CFP 2, DEM, Detroit, MI 48202, USA

Gerard Martin MD

Department of Emergency Medicine

Henry Ford Hospital CFP 2, DEM, Detroit, MI 48202, USA

doi:10.1016/j.ajem.2011.04.024

Potassium status should be evaluated also when diabetic ketoacidosis is complicated by heart failure^?

To the Editor,

The prevalence of hypokalemia quoted in the recent study [1] may not necessarily be a reflection of the state of

^? The author declares no conflict of interest.

affairs when Diabetic ketoacidosis coexists with hypervolemia attributable to heart failure. Documentation of potassium status in all body compartments is relevant to the management of these patients because unlike their hypovolemic counterparts who can tolerate the large volumes of intravenous fluids required to deliver appropri- ate doses of potassium replacement therapy [2], patients with DKA with coexisting heart failure and, hence, hypervolaemia incur the risk of worsening of heart failure with such treatment. On the other hand, if such treatment is withheld, there is a real risk of worsening of hypokalemia after the initiation of low-dose Intravenous insulin infusion. Although the prevalence of heart failure has not been quantified in patients with DKA, diabetes mellitus is a recognized risk factor not only for progression of asymptomatic left ventricular systolic dysfunction to symptomatic heart failure [3] but also for acute myocardial infarction (AMI) [4], heart failure being the outcome in most Diabetic patients with the latter complication [5]. Furthermore, as was the case in 6 of 258 diabetic subjects in the latter study, AMI can, itself, be a precipitating factor for DKA [5], an observation that resonates with the documentation of AMI in 11 of 257 subjects reported with DKA in 1 study [6].

Notwithstanding these considerations, intravenous fluids continue to be recommended as an adjunct to low-dose intravenous insulin even in the occasional patient with DKA with coexisting heart failure [7], the only caveat being the recommendation to administer intravenous fluids at a lower rate “in those patients with a history or signs of heart failure” [8]. A different approach was adopted in the management of 7 patients aged 76 to 91 years with coexisting clinically overt heart failure and diabetic decompensation, the latter characterized by pretreatment serum blood glucose in the range from 30.2 to 50.6 mmol/L, with concurrent serum potassium in the range from 4 to 5.7 mmol/L, serum urea in the range from 5 to 17 mmol/L, and serum bicarbonate in the range from 12 to 29 mmol/L (mean level, 22 mmol/L). In 6 of these 7 patients with coexisting heart failure and diabetic decompensation, low-dose intravenous Insulin infusion was administered without adjunctive intravenous fluid rehydra- tion therapy and, accordingly, without adjunctive intrave- nous potassium replacement.

Among the 6 patients who were not rehydrated with intravenous fluids was a 91-year-old woman with myocar- dial infarction and heart failure who was admitted with the following biochemical parameters: serum glucose,

50.6 mmol/L; serum potassium, 4.2 mmol/L; serum urea,

10.4 mmol/L; and serum bicarbonate, 16 mmol/L. Signs of cardiac failure included bilateral Pitting edema and elevation of jugular venous pressure, necessitating admin- istration of 40 mg intravenous frusemide. Thirteen and a half hours after initiation of low-dose insulin infusion, Serum glucose had fallen to 8.8 mmol/L, with concurrent serum potassium 3.0 mmol/L and concurrent serum bicarbonate 28 mmol/L. Serum potassium subsequently

increased to 3.2 mmol/L (with concurrent serum glucose,

11 mmol/L; serum urea, 13.1 mmol/L; and serum bicarbonate, 27 mmol/L) after administration of oral supplements of effervescent potassium.

In all 7 patients with coexisting heart failure and diabetic decompensation, the metabolic crisis was successfully resolved, as shown by the following posttreatment param- eters: mean serum glucose, 10.1 mmol/L (range, 8.2- 11.4 mmol/L); mean serum potassium, 3.8 mmol/L (range, 3.2-4.3 mmol/L); and mean serum urea, 11.0 mmol/L (range, 6.3-15.4 mmol/L) [9]. Posttreatment serum bicarbonate was documented in only 6 of the 7 patients, and it was recorded as 28 mmol/L (range, 24-32 mmol/L).

The management of potassium status is complicated by the risk of aggravation of fluid overload when DKA coexists with heart failure and by the risk of worsening of hypokalemia if potassium replacement is not delivered by the intravenous route. These issues are not addressed by studies of management of DKA that specifically exclude patients who have coexisting heart failure [2]. For a coherent treatment strategy to emerge, due recognition must be made not only of the prevalence of heart failure in DKA but also of the degree to which patients with DKA with heart failure are depleted of potassium in different compartments of body water.

Oscar M.P. Jolobe MB, ChB, DPhil

Manchester Medical Society c/o John Rylands University Library M13 9PP Manchester, United Kingdom

E-mail address: oscarjolobe@yahoo.co.uk doi:10.1016/j.ajem.2011.05.004

References

1. Arora S, Cheng D, Wyler B, Menchine M. Prevalence of hypokalemia in ED patients with diabetic ketoacidosis. Am J Emerg Med 2011doi: 10.1016/j.ajem.2011.01.002.
2. Umpierrez GE, Latif K, Stoever J, Cuervo R, Park L, Freire A, et al. Efficacy of subcutaneous insulin lispro versus continuous intravenous regular insulin for the treatment of patients with diabetic ketoacidosis. Am J Med 2004;117:291-6.
3. Das SR, Drazner MH, Yancy CW, Stevenson LW, Gersh BJ, Dries DL. Effects of diabetes mellitus and ischemic heart disease on the progression from asymptomatic left ventricular dysfunction to symp- tomatic heart failure: a retrospective analysis from the Studies of Left Ventricular Dysfunction (SOLVD) Prevention Trial. Am Heart J 2004;148:883-8.
4. Kannel WB, McGee DL. Diabetes and cardiovascular disease. The Framingham Study. JAMA 1979;241:2035-8.
5. Partman JO, Bradley RF. Acute myocardial infarction in 258 cases of diabetes. N Engl J Med 1965;273:455-61.
6. Beigelman PM. Severe diabetic ketoacidosis (diabetic coma) 482 episodes in 257 patients; experience of three years. Diabetes 1971;20: 490-500.
7. Page MMcB, Alberti KGMM, Greenwood R, Gumma KA, Hockaday TDR, et al. Treatment of diabetic coma with continuous Low-dose infusion of insulin. BMJ 1974;2(5921):687-90.
8. Basu A, Close CF, Jenkins D, Krentz AJ, Nattrass M, Wright AD. Persisting mortality in diabetic ketoacidosis. Diabet Med 1993;10: 282-4.
9. Jolobe OMP. Management of hyperglycaemic emergencies (letter). Proc R Coll Physicians Edinb 1995;5:338-9.

“Did scoop stretchers in spinal-injured patients study use an appropriate log-roll control group?”

To the Editor,

The authors of “Are scoop stretchers suitable for use on spine-injured patients?” are to be applauded for their innovative study and basic study design [1]. That said, the description of the methodology omitted an important detail regarding whether a specific variation of the Log-roll maneuver was used in the control group, and if the log roll was standardized, which technique was used.

This is not a trivial point because a prior study showed that spinal mobility varies between different log-roll techniques, as defined by the positioning of the patient’s arms [2]. Although not stressed in the provider manual, based on these differences, the Advanced Trauma Life Support program started recommending the “arms at side” technique in the Advanced Trauma Life Support Instructor Manual nearly 20 years ago [3]. Unfortunately, the relative lack of emphasis in provider publications has precluded effective standardization to the “arms at side” log roll, leading to log roll policies that are widely divergent or nonexistent.

If this real-world practice is reflected in the study’s control group, or worse, if the authors consistently used a technique other than “arms at side,” it could have provided a treatment bias in favor of the scoop stretcher group that would complicate generalization of the positive results reported by Del Rossi et al [1].

Robert E. Suter DO, MHA Division of Emergency Medicine UT Southwestern, Dallas, TX, USA E-mail address: resuter@gmail.com

doi:10.1016/j.ajem.2011.05.017

References

1. Del Rossi G, Rechtine GR, Conrad BP, Horodyski MB. Are scoop stretchers suitable for use on spine-injured patients? AJEM 2010;28(7): 751-6.
2. Suter RE, Tighe TV, Sartori J, Reed K. Thoraco-lumbar spinal instability during variations of the log roll maneuver. Prehosp Disaster Med 1992;7(2):133-8.
3. Personal Communication, American College of Surgeons Committee on

Trauma staff, 1993.