Key words
laparoscopy - intraoperative hypothermia - HumiGard
®
- pneumoperitoneum
Introduction
Intraoperative hypothermia is defined as a decrease in core body temperature to below
36 °C. Without the use of heat-preserving measures, the temperature of most patients
falls by 1 – 2 °C during an operation as their intrinsic thermoregulation switches
off [1]. Due to the effect of general anaesthesia, the body is no longer able to counteract
the reduction in temperature by shivering or vasoconstriction [1]. In addition, the room temperature is often low, and cooling takes place across
wound surfaces. Skin incisions for laparoscopy are much smaller than those of open
surgical procedures, but the entire internal abdominal body surface comes in contact
with the cold and dry insufflated gas (CO2). The temperature drop during laparoscopic procedures therefore does not differ essentially
from that seen during open surgery [2], [3]. Even mild forms of intraoperative hypothermia can lead to a marked increase in
morbidity and mortality. The negative consequences of intraoperative hypothermia have
been well researched and include disturbance of blood clotting with increased blood
loss and an increased rate of transfusions, myocardial dysfunction, arrhythmias and
hypokalaemia. In addition, delayed wound healing and wound infections occur more often,
with prolonged hospitalisation as a result [4], [5], [6].
CO2 at room temperature is usually employed in laparoscopic surgery for insufflation
of the pneumoperitoneum; it is relatively cold and dry compared with body temperature.
This factor is often overlooked when intraoperative hypothermia develops. Evidence
from studies conducted in pigs show, however, that the risk of hypothermia can be
reduced by the use of warmed and humidified insufflation gas [7], [8], [9], [10]. The question is particularly relevant in gynaecology as women in particular tend
towards intraoperative hypothermia [11].
The aim of this retrospective case control study was to investigate the effect of
humidified CO2 at body temperature on the intraoperative temperature profile in order to demonstrate
possible prevention of hypothermia.
Material and Methods
The retrospective, non-randomised case control study was conducted in the department
of gynaecology and obstetrics of RWTH Aachen University Hospital. Patients were allocated
to the study or control group using the independent variable “Use of HumiGard®”. All patients in whom HumiGard® had been used during their laparoscopy were assigned to the study group. This system
had not been used in the patients in the control group.
Patients
The study included patients who had undergone operative laparoscopy before 01.05.2011
and after 12.07.2013, as use of HumiGard® in the intervening period could not be traced from the documentation.
The patients selected for the control group were those who had surgery in RWTH Aachen
University Hospital after 04.01.2009 but before the HumiGard system was first used
(01.05.2011).
The study group comprised patients in whom the use of HumiGard® was documented after 12.07.2013.
A further inclusion criterion was documented use of the 3M™ Bair Hugger™ warming unit (3M, Neuss, Germany), which was part of the standardised heat management
of RWTH Aachen University Hospital.
The exclusion criteria were secondary laparotomy, repeated laparoscopies in the same
patient, planned operation duration less than 60 minutes, lack of documentation of
the intraoperative temperature, additional warming measures in the form of a warming
mat, warmed infusions or a combination of the two.
In the control group, the pneumoperitoneum was created and maintained using cold and
dry CO2 (room temperature and up to 20% humidity). In the study group, warm humidified CO2 was used (35 ± 2 °C, 98 ± 2% humidity). These parameters were achieved with the HumiGard
MR860 Surgical Humidification System® (Fisher & Paykel Healthcare Limited, Auckland, New Zealand).
The intra-oesophageal temperature was measured intraoperatively every 10 minutes.
The first temperature measured after the skin incision was defined as the start temperature.
The end temperature was the last measurement before release of the pneumoperitoneum
and wound closure.
110 patients were included, 59 in the study group and 51 in the control group. In
the control group, no other warming measures were used apart from a warm air blower.
There was complete documentation of the intraoperative temperature. Because of the
statistical case control matching, the analysis subsequently used only patients for
whom a statistical twin could be found (n = 33 per group) ([Fig. 1]).
Fig. 1 Inclusion and exclusion criteria. For the 110 patients included, 33 statistical pairs
of twins could be evaluated so 66 patients were used for statistical analysis.
Statistical analysis
To eliminate confounding factors from the statistical analysis, the study and control
groups were matched. In the case control matching, only the following two criteria
showed a Spearman correlation coefficient of > 0.15 with the key variable “start-end
temperature difference” and were therefore used as matching criteria: “operation duration”
and “smoker status (yes/no)”. Matching was performed by IBM® SPSS® Statistics 23.0.0. The operation duration tolerance range in which the two groups
were allowed to differ was set at five minutes. Patients to whom a statistical twin
from the other group could not be assigned were excluded from the study. To increase
the internal validity, a reduction in the case number from 110 to 66 was therefore
accepted.
The temperature data were examined for their statistical significance in an inductive
analysis. The case control matching resulted in two dependent samples so the differences
in the target variables between the respective twin pairs were examined for a normal
distribution. If this was present, a parametric procedure was used (paired t-test).
When the target variables were not normally distributed between the matched partners,
the Wilcoxon signed-rank test was used. The following characteristics were compared
between the two groups as target variables: age at the time of the operation, duration
of anaesthesia, number of previous operations, ASA classification, diabetes (yes/no),
main diagnosis, intraoperative infusions, hysterectomy operation technique, smoker
(yes/no), deeply infiltrating endometriosis (yes/no) and preparation time until operation.
Fig. 2 Temperature profile of the control and study groups, showing the temperature curves
using the start, middle and end of the operation. Measurement was intra-oesophageal.
Results
Demographic overview
66 patients were included in the analysis. Because of the case control matching, these
were divided equally between the two groups, resulting in a case number of 33 per
group. The 66 patients in the study were all women. The mean age was 43.6 years (± 11.9
years), ranging from 20 to 81 years. The median was 42 years.
Operations
Roughly half of all the operations were hysterectomies (34/66). While 7 patients had
laparoscopic supracervical hysterectomy (LASH), total laparoscopic hysterectomy (TLH)
was performed 27 times. The average time between start and end of surgery was 124.1
minutes (± 48.9 minutes); the shortest laparoscopy lasted 42 minutes and the longest
took 240 minutes. The median was 118.5 minutes.
Patient characteristics
There was no significant difference between the two groups for the patient characteristics
listed in [Table 1]. The study and control groups were therefore comparable with regard to the variables
listed in [Table 2].
Table 1 Comparison of the characteristics of the study groups. Group H: study group, group
0: control group
|
Variable
|
Group 0
|
Group H
|
p value
|
|
1 Paired t-test (normally distributed data)
2 Wilcoxon signed-rank test (non-normally distributed data)
|
|
Age at time of operation (years)
|
45
|
42
|
0.3131
|
|
Body Mass Index (kg/m2)
|
28
|
25
|
0.0791
|
|
Smoker (yes/no)
|
5
|
5
|
|
|
Diabetes (yes/no)
|
3
|
1
|
|
|
Number of previous operations
|
4
|
4
|
0.7151
|
|
ASA classification (1 – 5)
|
|
|
9
|
18
|
|
|
|
19
|
12
|
|
|
|
5
|
3
|
|
|
Main diagnostic categories
|
|
|
11
|
12
|
|
|
|
7
|
8
|
|
|
|
7
|
4
|
|
|
|
5
|
0
|
|
|
|
3
|
9
|
|
|
Deeply infiltrating endometriosis (yes/no)
|
2
|
3 [32]
|
|
|
Operation preparation time (min)
|
78
|
74
|
0.4761
|
|
Anaesthesia duration (min)
|
178
|
175
|
0.5421
|
|
Incision-suture time (min)
|
124
|
124
|
0.8672
|
|
Crystalloid solutions
|
1.36
|
1.61
|
0.2541
|
|
Colloid solutions
|
0.12
|
0.06
|
0.4231
|
Table 2 Start, middle and end temperature. The measurements were intra-oesophageal. Figures
in °C.
|
Start temperature
|
Middle temperature
|
End temperature
|
|
Control group
|
36.10 (± 0.46)
|
36.07 (± 0.42)
|
36.01 (± 0.40)
|
|
Study group
|
35.94 (± 0.46)
|
35.98 (± 0.49)
|
36.04 (± 0.49)
|
Temperature analysis
Comparison of the temperature profile in the two groups showed that the mean value
fell steadily in the control group, while it increased in the study group ([Fig. 2]). A test of significance was performed to confirm the opposite temperature profiles
([Table 3]). Both the start-end difference and, in individual sections, the start-middle and
middle-end difference were examined using the paired t-test. The start-end difference
was significantly greater in the study group at 0.09 °C than in the control group,
where it was − 0.09 °C (p = 0.011). No significance was found from the start to the
middle temperature (p = 0.122). In the second period examined, the middle-end difference
in the study group was highly significantly above that of the control group at more
than 0.11 °C (p = 0.003).
Table 3 Significance test of temperature differences at the three times: start, middle and
end temperature. The start-end, start-middle and middle-end differences were examined
by paired t-test. Statistical significance was assumed at p < 0.05.
|
Control group
|
Study group
|
t
|
df
|
p
|
|
df: degrees of freedom
|
|
Start-end difference
|
− 0.09
|
0.09
|
2.70
|
32
|
0.011
|
|
Start-middle difference
|
− 0.03
|
0.04
|
1.59
|
32
|
0.122
|
|
Middle-end differences
|
− 0.06
|
0.05
|
3.21
|
32
|
0.003
|
In both groups, the percentage of intraoperative hypothermia at the start and end
of the operation was examined. In the study group, the proportion of patients with
hypothermia fell markedly from more than 54.22 to 36.36%. By contrast, the proportion
of patients with hypothermia increased from 36.36 to 42.42% in the control group ([Table 4]).
Table 4 Percentage of hypothermic patients at the three measurement times (start, middle
and end temperature). Figures in percent.
|
Hypothermia (in %)
|
Group 0
|
Group H
|
|
Percentage of start temperatures
|
36.36
|
54.55
|
|
Percentage of middle temperature
|
39.39
|
42.42
|
|
Percentage of end temperatures
|
42.42
|
36.36
|
Discussion
The decrease in temperature during laparoscopic surgery is due to three causes: the
effect of general anaesthesia, environmental factors and the insufflation of dry gas
at room temperature. General anaesthesia accounts for most of the decrease as it paralyses
the bodyʼs intrinsic thermoregulation and reduces metabolic heat production by about
30%. In addition, the room temperature in the operating theatre is often low. The
HumiGard® system focuses on the third cause: temperature loss due to insufflation of cold dry
gas. This undergoes secondary warming and humidification in the abdominal cavity until
it reaches the temperature and humidity of the abdomen and equilibrium is achieved.
The heat and also the water content of the peritoneum are diminished continuously
during operation. This loss of energy can contribute to the development of hypothermia
[7]. Since the peritoneal surface area roughly corresponds to the outer body surface,
the importance of attempting to protect intra-abdominal heat becomes clear.
According to Sessler (2016), the core body temperature can provide the most reliable
information about a patientʼs thermal status. The following four sites are suitable
for measuring it accurately: the nasopharynx, the distal oesophagus, the ear drum
by means of a contact thermometer and the pulmonary artery. The data used in our analysis
are based on measurements in the nasopharynx and accordingly deliver valid information
on core body temperature [1].
This study showed that use of HumiGard® can prevent a decrease in body temperature. The temperature profile curves ran in
the opposite direction to those of the control group. In the latter, there was a drop
in body temperature especially in the second half of the operation. The HumiGard® system, which is the only one of its kind on the European market, reduced the rate
of hypothermia steadily. By contrast, the rate of hypothermia rose steadily in the
course of the operation in the control group. The high preoperative percentage of
hypothermia of approximately 55% in the HumiGard group is noteworthy. Possible reasons
are a long interval of time between induction and the actual start of operation, as
well as inadequate or omitted prewarming. However, the mean preparation time until
operation differ only minimally between the two groups so prewarming appears a more
likely explanation. Possible differences in the preoperative procedure at the time
of the historical controls may be the reason for the lower percentage of hypothermia
at the start of the operation in this group. Effective prewarming results in an intraoperative
body temperature that is 0.4 °C higher compared with controls [1]. However, the HumiGard system reduced hypothermia steadily compared with the control
group with a warm air blower only.
In animal studies, in which exclusive warming was investigated primarily, no significant
effects were found for avoiding hypothermia. In these animal studies, however, complex
heat management (e.g. a warm air blower) was not used. The average incision-to-suture
time in these studies was over 120 minutes [9], [12], [13].
The positive effects of conditioned gas on human temperature metabolism were proclaimed
in the 1998 publication by Ott et al. on warmed and humidified carbon dioxide. This
research team used the Insuflow® system, which is similar to the HumiGard® system and is licensed in the United States. This prospective randomised multi-centre
study, which used additional heat management for both groups, showed a temperature
difference in favour of Insuflow®, which reduced the rate of intraoperative hypothermia [14]. Another recently published randomised study, which additionally used a warm air
blower and warm infusions, did not find any significant temperature difference with
the use of HumiGard®. However, this study cannot be used for comparison as the operation times were very
short (average of 15 minutes insufflation time [15].
By contrast, we observed the significant temperature differences only in the second
half of the operation. In our study, the planned minimum duration of all surgical
procedures was 60 minutes.
Studies in humans, which included bariatric patients having laparoscopic reduction
of stomach size, did not find any positive effect on intraoperative temperature maintenance
[16], [17]. It should be noted, however, that overweight patients generally tend less to intraoperative
hypothermia due to the insulating effect of fat tissue [18].
Consistent with our retrospective case control study, a recent Cochrane analysis in
2016 showed that statistically significantly higher core temperatures were found when
warmed and humidified insufflation gas was used. Since the authors criticise the quality
of the previous clinical studies, the results will have to be investigated in further
studies [19].
Although our study is a retrospective analysis, its strength lies particularly in
the precise matching of the patients, the operation duration and the employed warming
method. There is therefore high internal validity due to the high comparability of
the two groups of patients.
These results show that external body warming by a warm air blower can compensate
the initial drop of 1 – 2 °C in body temperature due to general anaesthesia. Other
options consist of use of warming blankets and also the use of warmed infusions. An
adhesive plastic sheet is used more often in open surgery. Ultimately, however, these
measures do not affect the peritoneum-lined body cavity. Especially during more protracted
laparoscopic operations, additional warming and humidification of the insufflation
gas can lead to better maintenance of body temperature and avoidance of hypothermia
in the second half of the operation and should therefore be used in addition. The
authors of a recently published review also come to this conclusion [20].
Accordingly, there are conflicting publications regarding the question we investigated,
sometimes with opposite conclusions. Previous studies are difficult to compare, however,
as the patient populations were heterogeneous, the use of additional warming methods
differed, the operation times differed markedly, and temperatures were also measured
in various ways.
Even though the focus of the present study was on the intraoperative temperature profile
and the occurrence of any hypothermia, it should be noted that further beneficial
effects are associated with the use of warmed and humidified laparoscopy gas: reduction
of postoperative wound infections, postoperative pain and postoperative analgesic
consumption have been described. Because of this, the duration of patient hospitalisation
can be reduced, which is of great economic significance [6], [10]. Avoidance of mesothelial damage due to drying of the peritoneum can prevent the
consequent inflammatory reaction with the development of a fibrin matrix, which acts
as the basis for adhesions [9].
A precise evaluation of the aforementioned effects, especially from the economic aspect,
is still the object of research and must be studied in larger numbers of patients.
Conclusion
An overall tendency to positive assessment of warming and humidification of insufflation
gas is apparent in the literature. By excluding as many confounding factors as possible,
the present study delivers the important finding that warming and humidification of
the insufflation gas CO2 can make an important contribution to intraoperative temperature maintenance, including
avoidance and reduction of intraoperative hypothermia. The small case number due to
the strict inclusion and exclusion criteria and the retrospective study design are
disadvantages, which will be put into perspective in the prospective randomised clinical
TePaLa study (Temperature and Pain in Laparoscopy, ClinicalTrials.gov Identifier:
NCT02781194) and other clinical studies.
