• Synopsis
• Reference
• Story
• Protocol: Austrian study
• Protocol: Johns Hopkins study
• Results: Austrian study
• Table 1: Characteristics of the Patients in the Two Study Groups
• Table 2: Postoperative Findings in the Two Study Groups
• Table 3: Postoperative Findings in the Study - Patients According to Smoking Status
• Results: Johns Hopkins study
• Table 1: Intraoperative Cardiac Outcomes
• Table 2: Postoperative Cardiac Outcomes
• Table 3: Intraoperative Patient Characteristics
• Questions
• Definitions
• Credits

Reference

Kurz, A., Sessler, D. I., and Lenhardt, R. (1996). "Perioperative Normothermia to Reduce the Incidence of Surgical-Wound Infection and Shorten Hospitalization," The New England Journal of Medicine, 334 (19), 2109-2115.

Frank, S., Fleisher, L., Breslow, M., Higgins, M., Olson, K., Kelly, S., and Beattie, C., (1997). "Perioperative Maintenance of Normothermia Reduces the Incidence of Morbid Cardiac Events," Journal of the American Medical Association, 277 (14), 1127-1134.

When patients undergo surgery, the operating room is kept cool so that the physicians in heavy gowns will not be overheated. The price for the surgeons' comfort could be paid by the patient. The exposure to cold, in addition to impairment of temperature regulation caused by anesthesia and altered distribution of body heat, may result in mild perioperative hypothermia (approximately 2°C below the normal core body temperature). As a result of the hypothermia, patients may have an increased susceptibility to perioperative wound infections or even morbid cardiac events.

In Austria, Kurz, Sessler, and Lenhardt (1996) investigated whether maintaining a patient's body temperature close to normal by heating the patient during surgery decreases wound infection rates. In a separate study at Johns Hopkins, Frank, et al. (1997) examined whether maintaining a patient's body temperature close to normal during surgery is associated with fewer incidents of (morbid) heart attack.

[TOP]

Researchers Kurz, Sessler, and Lenhardt (1996) used the following protocol for their study:

All patients included in the study were undergoing colon or rectal surgeries, which are typically associated with a high risk of infection. The night before surgery, all patients were prepared for the next day's surgery in a standard manner. Both during and after surgery patients were given liquids intravenously in hopes that this would decrease the chance of infection caused by bacteria from the wound entering the bloodstream. Random assignment of patients to the temperature management groups was done during the induction of anesthesia. Codes which had been generated by computer and numbered and sealed in opaque envelopes were used for the random assignments. The two groups were the normothermic group (patients' core temperatures were maintained at near normal 36.5°C) and the hypothermic group (patients' core temperatures were allowed to decrease to about 34.5°C). While intravenous fluids were administered through a fluid warmer machine for both groups, only the normothermic group had the warmers activated. Additionally, a forced-air cover was used on the upper bodies of patients in both groups, but it delivered heated air in the normothermic group only (it delivered surrounding air in the hypothermic group). In order to keep the surgeons and operating room personnel from detecting which patient was in which group, shields and drapes were placed over all devices which would indicate group assignment. Postoperatively there was no controlling of patient temperatures, and the patients did not know their group assignments. All patients were able to self-administer pain killers.

Determinations of when to begin postoperative feeding, suture removal, and hospital discharge were made by attending surgeons who were unaware of the patients'; group assignments and core temperatures during surgery. Routine surgical considerations were involved in the discharge decisions. Note that the study was done in Austria, a country with managed health care, so that there are no insurance or administrative issues that might contribute to patient-to-patient variation in the length of stay at the hospital.

[TOP]

The subjects in the study were 300 patients undergoing abdominal, thoracic, or vascular surgery, who either had documented coronary artery disease or were at high risk for coronary disease. These patients were randomly assigned to receive routine thermal care (hypothermic care) or normothermic care. According to Frank, et al. (1997),

…routine thermal care. . . was delivered according to the following protocol. The thermostat in the operating room was set to approximately 21°C. Intravenous fluids and blood were warmed [with the same type of device]. A [specific] heat-moisture exchanger. . . was used in the respiratory circuit for patients receiving general anesthesia. Intraoperatively, the patient was covered above and below the field [i.e., location of surgery] with one layer of paper surgical drapes. Postoperatively, either one or two warmed cotton blankets were placed over the patient, at the nurse's discretion.

Patients in the normothermic group were treated as follows. . . . The thermostat in the operating room was set to approximately 21°C, fluids and blood were warmed, and a heat-moisture exchanger was used in the respiratory circuit. Depending on the surgical site, an upper- or lower-body forced-air warming cover was placed over the patient. During the intraoperative and postoperative periods, both the temperature and airflow settings were adjusted to maintain core temperature at or near 37°C.

Since temperature was monitored via probes on the arm and fingertips during the postoperative period, the forced-air blanket was placed over the patient's legs and trunk, leaving the monitored arm exposed.

[TOP]

The following are excerpts from tables which were included in the Kurz, Sessler, and Lenhardt (1996) paper. For categorical variables, the tables give the number (and/or percent) of patients; for numerical variables, the tables give the mean ± (plus or minus) one standard deviation.

Table 1: Characteristics of the Patients in the Two Study Groups

The following are excepts from tables which appeared in the Frank, et al. (1997) paper. For categorical variables, the tables give counts and percentages; for continuous variables, the tables give mean ± one standard error for the mean.

[TOP]

Questions

Question 1)
a) Was the New England Journal study

(i) a designed, controlled experiment or

(ii) an observational study?

b) Which of the following might introduce bias into this study? Explain your choice.

(i). One surgeon performs all surgeries on the normothermic group while another performs all surgeries on the hypothermic group.

(ii). For each patient, a fair coin is tossed to determine which surgeon will operate on the patient.

c) Why did the researchers use double-blinding in this study?

Question 2) Review the protocol of the Johns Hopkins study.

a) What variables or (heating) factors are controlled in the Johns Hopkins study?

b) Which factors differ between the treatment and control groups in the Johns Hopkins study?

c) The protocol allows for patients in the control group to be given either one or two warmed cotton blankets, at the nurse's discretion. Explain why the researchers may have chosen to allow such subjectivity in the protocol of the study.

Question 3) One of the main points in the article by Kurz, Sessler, and Lenhardt (1996) was that the warming blankets (along with the other warming strategies), used for the normothermic group to keep them at or near normal body temperature during surgery, reduced the chance of infection. One effect of this would be that the patients in the normothermic group (the warming group) should have shorter hospital stays than those in the hypothermic group. To examine this, do the following:

a) Using the data from Table 2, perform an appropriate significance test for determining whether the treatment (use of warming blankets) affects prevalence of postoperative infection.

b) Using the data for all patients from Table 2, construct a 95% confidence interval for the difference between the means for length of stay in the hospital for the normothermia and hypothermic groups. What does this interval tell you about the effect of the treatment?

c) One might conclude, based on the results found in Parts a and b, that the use of warming blankets during surgery reduces infection rates and, therefore, decreases length of stay in the hospital. However, the use of the warming blanket might decrease length of stay for some reason unrelated to infection. Using the data from Table 2, carry out the significance test to determine whether the treatment affects length of stay for uninfected patients. Compute the test statistic, find the p-value, and state your conclusions.

Question 4) In the Results section for the Johns Hopkins study, the table of intraoperative patient characteristics gives the mean plus or minus (±) the standard error of the mean for various characteristics.

After reviewing that table, several students from an introductory Statistics class engage in the following debate. Evaluate their comments and decide which, if any, of the arguments are correct.

Student 1: About 95% of the surgeries for patients in the hypothermic group lasted between 1.2 and 5.6 hours. In the normothermic group, about 95% of the surgeries lasted between 1.8 and 5.4 hours.

Student 2: A 95% confidence interval for the mean duration of surgery for hypothermic patients is = (3.31, 3.49). For normothermic patients the interval is = (3.52, 3.68).

Student 3: The standard error for the mean seems unusually large, since the standard deviation for the individual observations is equal to the standard error of the means multiplied by the square root of n. Thus the mean plus or minus (±) one standard deviation would go far outside the range of possible values for duration of surgery.

Question 5) Kurz, Sessler, and Lenhardt (1996) stated that "those who smoked had three times more surgical-wound infections and significantly longer hospitalizations than the nonsmokers."

a) What do you think the researchers are referring to?

b) The hypothermic group had three times more infections than the normothermic group. Do you think, therefore, that smoking (rather than the treatment) may be the reason for the difference in the number of infections between the normothermic and hypothermic groups?

Question 6) Why did the researchers choose to make all of the comparisons listed in Table 1?

Question 7) Recall that the patients studied by Kurz, Sessler, and Lenhardt (1996) were adults undergoing colon or rectal surgery. Do the results we find in this data carry over to:

• other types of surgeries?

• shorter/longer surgeries?

• surgeries on children?

[TOP]

Definitions

perioperative: around the time of surgery

hypothermia: abnormally low body temperature

morbid: pertaining to, arising from, or affected by disease.

psychosomatic: relating to a disorder having physical symptoms but originating from mental or emotional causes.

Normal body temperature is around 37.0°C (98.6°F)

[TOP]

Thanks to Rusch International for supplying a picture of their surgical blanket.