Inducing therapeutic hypothermia using chilled saline in resuscitated cardiac arrest patients has been shown to be feasible and effective. Limited research exists assessing the efficiency of this cooling method. The objective of this study was to assess the change in temperature of 4°C saline upon exiting an infusion set in the laboratory setting while varying conditions of fluid delivery.
Efficiency was studied by assessing change in fluid temperature (°C) during the infusion under four laboratory conditions. Each condition was performed four times using 1-L bags of normal saline. Fluid was infused into a 1000-mL beaker through 10 gtt/mL tubing. Flow rate was controlled using a tubing clamp and in-line transducer with a flow meter, while temperature was continuously monitored in a side port at the terminal end of the intravenous (IV) tubing using a digital thermometer. The four conditions included different insulation methods. Descriptive statistics and analysis of variance were performed to assess changes in fluid temperature.
The mean (±SD) fluid temperature at time 0 was 3.2°C (95% confidence interval [CI] = 3.0 to 3.4 °C) with no significant difference in starting temperature between groups (p = 0.45). When flow rate was constant, it was determined that fluid temperatures were significantly cooler when infused using a chilled, gel-filled sleeve around the saline bag (p < 0.006).
In a laboratory setting, the most efficient method of infusing cold fluid appears to be a method that both keeps the bag of fluid insulated and infused at a faster rate.
Evaluacion de la Influencia de un Aislante en la Temperatura de Infusión de Líquidos Intravenosos
Se ha documentado que la inducción de hipotermia terapéutica mediante el uso de suero salino frío en los pacientes recuperados de una parada cardiaca es factible y efectiva. Existen escasas investigaciones que valoren la eficiencia de este método de enfriamiento. El objetivo de este estudio fue valorar el cambio en la temperatura del suero salino a 4°C al salir de un aparato de infusión en el escenario de un laboratorio mientras variaban las condiciones de administración del líquido.
Se estudió la eficiencia valorando el cambio en la temperatura del líquido (ºC) durante la infusión en cuatro circunstancias experimentales. Cada circunstancia se realizó cuatro veces usando bolsas de 1 litro de suero salino. El líquido se infundió en un recipiente de 1000 mL a través de un tubo a 10 gotas/mL. El flujo se controló mediante una abrazadera en el tubo y un transductor en línea con un caudalímetro, mientras la temperatura fue continuamente monitorizada en la parte final de la conexión del tubo intravenoso por medio de un termómetro digital. Las cuatro circunstancias incluyeron diferentes métodos de aislamiento. Se realizó estadística descriptiva y análisis de la varianza para valorar los cambios en la temperatura del líquido.
La media de temperatura del líquido en el tiempo 0 fue de 3,2 °C (IC 95% = 3,0 a 3,4 °C) sin diferencias significativas en la temperatura inicial entre los grupos (p = 0,45). Cuando el flujo fue constante, se determinó que las temperaturas del líquido fueron significativamente más frías cuando se infundió usando un instrumento refrigerador, una funda de gel fijado alrededor de la bolsa del suero salino (p < 0,006).
En el escenario de un laboratorio, el método más eficiente de infusión de líquido frío parece ser un método que mantiene tanto la bolsa del líquido aislada como el ritmo de infusión más rápido.
Several studies have investigated the effectiveness of chilled saline for the induction of therapeutic hypothermia; however, there is scant research describing the most efficient method for delivering chilled saline.[1-7] If induction of therapeutic hypothermia using chilled saline is to become a standard practice, then identifying a practical technique that will allow fluid to be infused closest to its initial refrigerator temperature is important. Therefore, the efficiency of infusion techniques should be measured and tested. The objective of this study was to assess the change in temperature of chilled saline upon exiting an infusion set in the laboratory setting while varying certain conditions. It was hypothesized that fluid temperature exiting an infusion set would vary dependent on insulation of the saline bag and infusion set.
This study occurred in the Carolinas Medical Center Emergency Medicine Research Laboratories. This study did not involve human or animal subjects; no data were obtained through intervention or interactions with individuals.
Changes in fluid temperature were assessed over four different conditions where the rate of fluid flow was held constant (105 mL/min) with varied methods of infusion-set insulation. Four replicate tests were conducted for each of the four conditions in a room with a consistent temperature of 23°C. Each test was performed using prechilled 1-L bags of normal saline and by the same three investigators. All fluid, infusion set tubing, and insulation materials were stored in a walk-in refrigerator (Environmental Specialties Inc., Raleigh, NC) that was maintained at a temperature between 2 and 3°C. Prior to initiating a test, one of the investigators would step into the refrigerator, select a bag of fluid, and insert a temperature probe into the infusion set tubing port. Once the temperature probe was placed, the investigator would leave the refrigerator with the tubing, start a 6-minute countdown timer, and walk back to the laboratory. Upon arriving at the laboratory the bag of fluid with temperature probe in place was placed into an empty sink until 6 minutes had elapsed. This 6-minute waiting time was used to approximate a real-life waiting period from patient contact to initiating intravenous (IV) access during resuscitation as reported in an earlier randomized trial conducted by the study authors.
After the 6-minute waiting period had ended, the temperature of the saline was recorded (time 0), the infusion set tubing was connected, and the infusion flow rate was set. Fluid was infused into a 1-L beaker placed in the bottom of a sink. Temperature measurements were recorded for every 100 mL of fluid infused into the beaker. The test was terminated after 1000 mL of fluid was observed in the beaker (Data Supplement S1, available as supporting information in the online version of this paper).
The first (control) condition tested was a controlled infusion of chilled saline with no insulation. The next condition consisted of coiling and placing 3 feet of infusion set tubing between gel packs cooled to 3°C; extension tubing was not added to the infusion set (Data Supplement S2, available as supporting information in the online version of this paper). The third condition consisted of constructing an insulated and chilled sleeve in which to place the saline filled bag. A standard IV pressure infusion sleeve was injected through an existing air valve, with 250 mL of cooling gel (Quintiles Laboratories, Durham, NC). This amount of cooling gel was chosen as it was circumferential, and the most gel that could be injected into the sleeve while still allowing the bag of saline to fit. The gel used was proprietary and not developed by the investigators. The gel packs are typically used to keep specimens cool during shipping. The prechilled saline bag was placed inside the gel-filled sleeve while inside the refrigerator, before beginning the infusion experiment. The final method tested included using a gel-filled sleeve in conjunction with the infusion tubing coiled between cooled gel packs (a combination of the second and third conditions) to determine if there was additive benefit from combining both methods of insulation.
Flow rate was measured in mL/minute using an in-line transducer (N-series, Transonic Systems Inc., Ithaca, NY) with a flow meter (T206, Transonic Systems Inc.) and controlled using a tubing clamp. The flow meter was placed proximal to the last side port on the infusion set. Temperature was continuously monitored in a side port at the terminal end of the IV tubing using a digital thermometer (4600 Precision Thermometer, Yellow Springs Instruments, Yellow Springs, OH) displaying degrees Celsius (Data Supplement S3, available as supporting information in the online version of this paper). Volume of fluid infused was observed in a 1-L beaker to determine when measurements were to be recorded.
Data analysis was primarily descriptive, reporting mean temperatures with 95% confidence intervals (CIs) for each 100 mL of fluid infused per test. These results are reported with point estimates and 95% error bars. Between- and-within group comparisons were conducted using analysis of variance with a repeated-measures analysis performed when assessing the effect of group on temperatures. In this analysis, the overall probability of making a type I error was set at 0.05. To reduce the probability of a type I error, Bonferroni correction was applied to both stages of this analysis. Therefore, the alpha level for each individual test was set at 0.016.
The mean (±SD) fluid temperature at time 0 was 3.2°C (95% CI = 3.0 to 3.4° C), with no significant difference in starting temperature between groups (p = 0.45). Figure 1 compares infusion temperatures between the control condition (noninsulated infusion) at 105 mL/min and the two separate insulation conditions (see Methods), coiled tubing between cooled gel packs and the gel-filled sleeve. Figure 1 indicates that increases in fluid temperature were attenuated in the gel-filled sleeve group. The mean (±SD) fluid temperatures after 100 mL had been infused were 7.35 (±SD)°C (95% CI = 6.91 to 7.79°C; 105 mL/min), 6.15 °C (95% CI = 5.69 to 6.61°C; cooled tubing), and 6.95 °C (95% CI = 6.45 to 7.44°C; gel-filled sleeve). The mean (±SD) fluid temperatures after 1,000 mL had been infused were 11.40 °C (95% CI = 11.18 to 11.62°C; 105 mL/min), 11.32 °C (95% CI = 10.99 to 11.66°C; cooled tubing), and 7.75 °C (95% CI = 7.55 to 8.00°C; gel-filled sleeve). The overall test of significance indicated that there was a difference in fluid temperatures between groups (p = 0.002). Pairwise comparisons indicated that there was a significant difference in temperature when comparing the 105 mL/min group to the gel-filled sleeve group (p = 0.006). There was no statistical difference between the cooled tubing group and the gel-filled sleeve group (p = 0.032). However, at 700 mL, the 95% confidence bars of the cooled tubing group and the gel filled sleeve group no longer overlapped. When comparing the gel-filled sleeve with the combination of gel-filled sleeve and cooled tubing (condition 4), pairwise comparisons indicated that there was no significant difference in fluid temperatures between these groups (p = 0.83).
Results from this study demonstrated that insulation can attenuate changes in the temperature of chilled saline as infused in a laboratory setting. The model that infused fluid at a flow rate of 105 mL/min and used a cooled, gel-filled sleeve surrounding the saline bag appeared to be the most efficient method for maintaining fluid temperature as close to time 0. While the model that included wrapping the infusion tubing in a gel pack with or without the gel-filled sleeve was not significantly different from the gel-filled sleeve alone, practical differences may make this approach less effective. Restricting the amount of free tubing between the bag of fluid and the patient may not always be easily achieved, while placing the fluid bag in a gel-filled sleeve may be more likely to occur on a regular basis. The lack of a significant difference between the cooled tubing and the gel-filled sleeve may be explained by the fact that the fluid temperatures do not appear to significantly diverge from each other until after 600 mL of fluid had been infused.
Future translational work using the concepts tested should be performed to determine if these laboratory models affect induction time and eventually outcome among patients requiring therapeutic hypothermia during postarrest resuscitation. If the rapid induction of hypothermia is a clinically important goal, further studies are needed to assess the most efficient and effective method of temperature control during induction.
While the study investigators who performed these tests were not blinded to condition, in an attempt to limit bias, standardized work was used where each investigator performed the same tasks during each run of each condition. Further assessment of the feasibility of the described interventions is warranted. If out-of-hospital resuscitation is taken into consideration, results from this study may vary due to the different environmental conditions present in those resuscitations.
This study was conducted in a temperature-controlled laboratory; it is likely that fluid temperatures may increase more rapidly if the environment is warm, such as outside or in an un–air-conditioned house. For the in-hospital setting, translational research starting with animal models is likely warranted to assess the effect of these interventions on cooling times and mortality.
Results of this study indicate that fluid temperature exiting an infusion set varied significantly dependent on insulation of the saline bag and infusion set.