Abstract
Purpose.:
To determine how different intraoperative surgical procedures affect the midvitreous temperature.
Methods.:
The vitreous temperatures were monitored continuously with an intravitreal thermocouple in 87 eyes of 81 cases undergoing vitrectomy. Thirty-three eyes had diabetic retinopathy (DR), 35 eyes had an epiretinal membrane, and 19 eyes had an idiopathic macular hole. In eyes with DR, the correlation between the number of photocoagulations (PCs) and the change in temperature was analyzed. The temperature was also recorded before and after combined phacoemulsification and aspiration (PEA) and vitrectomy in 10 eyes.
Results.:
The average midvitreal temperature before the vitrectomy was 33.0 ± 1.3°C, 30.7 ± 1.7°C after core vitrectomy, 32.9 ± 1.3°C after membrane peeling, and 29.2 ± 1.4°C after peripheral vitrectomy. The temperature before PC was 29.8 ± 1.3°C, and it increased to 31.5 ± 1.9°C post-PC. The differences in the temperatures between consecutive procedures were significant (P < 0.01, respectively, Wilcoxon signed-rank test). The difference in the temperatures of the same procedures among the different diseases was not significant except after membrane peeling. A significant correlation was detected between the number of PCs and the duration of the PCs, and between the duration of PCs and the change in vitreous temperature after PC (r = 0.719, P = 0.0010, and r = 0.800, P = 0.0002, respectively, Spearman's rank correlation coefficient test). The temperature after PEA decreased significantly by 2.3°C.
Conclusions.:
Our results showed that vitreous temperatures vary during different vitrectomy procedures.
We studied 87 eyes of 81 consecutive patients undergoing vitrectomy by a single surgeon (HT) between October 2009 and June 2011 at the Nagoya University Hospital. The procedures used for this study were approved by the Ethics Committee of the University Hospital, and an informed consent was obtained from all patients after an explanation of the procedures to be performed. All of the procedures conformed to the tenets of the Declaration of Helsinki.
Forty-seven of the patients (51 eyes) were men and 34 patients (36 eyes) were women. The average age of the patients was 62.8 ± 11.6 years with a range from 28 to 83 years. Thirty-three eyes of 29 patients had diabetic retinopathy (DR), 10 eyes of seven patients had diabetic macular edema (DME) in eyes with proliferative diabetic retinopathy (PDR), and 23 eyes of 22 patients underwent vitrectomy for PDR. Thirty-five eyes of 34 patients had an epiretinal membrane (ERM), and 19 eyes of 18 patients had an idiopathic macular hole (MH).
The vitreous temperature was monitored intraoperatively during the entire surgery until the time of fluid–air exchange in all 87 eyes. The temperature recorded just after each procedure was used to represent the temperature during that procedure (
Table 1). The duration of the procedures was obtained retrospectively from the charts and used for the analyses.
Table 1 The Vitreous Temperature in Each Procedure
Table 1 The Vitreous Temperature in Each Procedure
Vitrectomy was performed after retrobulbar anesthesia by 5 mL of 2% lidocaine + 0.5% xylocaine. A 23-gauge or 25-gauge of Accurus Vitrectomy System (Alcon, Fort Worth, TX, USA) was used for the vitrectomy, and 68 eyes underwent combined phacoemulsification and aspiration (PEA) and vitrectomy and 19 eyes underwent vitrectomy alone. Six eyes were pseudophakic, and 13 eyes were left phakic.
The vitreous temperature was monitored throughout vitrectomy with an MLT 1402 T-type Ultra Fast Thermocouple Probe (Power Lab; AD Instruments, Castle Hill, Australia), which is a tissue-implantable microprobe with an ultrafast response time of 0.005 seconds. The sensor is installed at the tip of a 0.22-mm-diameter wire, and the wire was inserted into the midvitreous through a 30-gauge scleral incision created by a disposable 30-gauge needle. The thermocouple was inserted through the sclerotomy site, 3.5 mm from the corneal limbus and at the same clock hour midway between the medial rectus muscle and inferior rectus muscle insertions.
An infusion cannula and two trocars were placed at the conventional sites, and standard 23-gauge or 25-gauge vitrectomy was performed. After the core vitrectomy was completed, a posterior vitreous detachment was created if one was not already present. The ERM and internal limiting membrane in eyes with ERM, MH, and DME, and the fibrovascular membranes in eyes with PDR were removed, followed by peripheral vitrectomy with a wide angle viewing system and scleral indentation. Triamcinolone acetonide was injected into the vitreous of all eyes to make the internal limiting membrane more visible. Photocoagulation (PC) was performed on all eyes with a retinal tug, retinal tear, or lattice degeneration at the equator in cases with an ERM or MH and in all cases with DR. Then, fluid–air exchange was performed in 26 eyes to secure wound closure. Perfluoropropane (C3F8) was injected into all 19 eyes with a MH, and silicon oil was injected into seven eyes with PDR.
The Kruskal-Wallis test was used to determine the significance of differences between the vitreous temperatures for the three diseases (
Table 1). The Wilcoxon signed-rank test was used to determine whether there were significant differences in the changes of the temperatures between the different procedures (
Table 1).
The Mann-Whitney's
U test was used to determine the significance of the differences in the temperature (
Table 2).
Table 2 The Temperature Before and After Cataract Surgery
Table 2 The Temperature Before and After Cataract Surgery
| Temperature |
| Case No. | Before Cataract Surgery, °C | After Cataract Surgery, °C | Difference in Temperature, °C |
| 1 | 33.5 | 30.7 | 2.8 |
| 2 | 34.0 | 29.4 | 4.6 |
| 3 | 32.3 | 29.8 | 2.5 |
| 4 | 34.4 | 31.5 | 2.9 |
| 5 | 34.6 | 32.4 | 2.2 |
| 6 | 34.4 | 32.7 | 1.7 |
| 7 | 34.2 | 32.3 | 1.9 |
| 8 | 34.2 | 32.5 | 1.7 |
| 9 | 34.2 | 32.5 | 1.7 |
| 10 | 34.3 | 33.2 | 1.1 |
| Mean ± SD | 34.0 ± 0.7 | 31.7 ± 1.3 | 2.3 ± 1.0 |
The correlations between the number of PCs and the changes in the vitreous temperature, and the correlation between the number of PCs and the duration of PCs (n = 22) in eyes with DR were analyzed by Spearman's rank correlation coefficient. The correlations between the duration of membrane peeling and the change in vitreous temperature in eyes with DR (n = 18) were analyzed by Spearman's rank correlation coefficient test.
The correlations between the change in vitreous temperature and the duration of core vitrectomy (n = 51), and the correlation between the change in vitreous temperature and the duration of peripheral vitrectomy (n = 38) were analyzed by Spearman's rank correlation coefficient test.
The temperature before and at the end of each procedure is presented as the mean ± SD (
Table 1,
Fig. 1). The room temperature was 26.4 ± 0.8°C, the body temperature at the beginning of surgery was 36.3 ± 0.4°C, and the temperature of the irrigating solution at the beginning of surgery (BSS-plus [Balanced Salt Solution-plus]; Alcon) was 25.8 ± 1.4°C. The vitreous temperature just before vitrectomy was 33.0 ± 1.3°C. During core vitrectomy, the vitreous temperature decreased by 2.3 ± 1.6°C to 30.7 ± 1.7°C. During membrane peeling, the temperature gradually increased and reached 32.9 ± 1.3°C at the end of the procedure. During peripheral vitrectomy, the temperature decreased to 29.2 ± 1.4°C, which is 3.8 ± 1.6°C lower than the temperature just before vitrectomy. These changes in vitreous temperature between consecutive procedures were significantly different (
P < 0.01, Wilcoxon signed-rank test). The temperature increased to 31.5 ± 1.9°C during PC, which is 1.7 ± 2.0°C higher than that before PC.
The number of PC applications was 2153 ± 974 in eyes with DR (n = 33), 281 ± 369 in eyes with an ERM (n = 15), and 454 ± 669 in eyes with a MH (n = 13).
The differences in the vitreous temperature during each procedure such as just before vitrectomy, after core vitrectomy, after peripheral vitrectomy, before PC, and after PC among the different diseases were not significant (P > 0.05; Kruskal-Wallis test).
The changes in vitreous temperatures before and after PC in eyes with DR are shown in
Figure 2. The vitreous temperature increased after PC, and there was a significant correlation between the number of PCs and the change in vitreous temperature (
r = 0.578,
P = 0.0011;
n = 33; Spearman's rank correlation coefficient test).
The number of PCs in eyes with DR was significantly correlated with the duration of the PC (13.9 ± 6.1 minutes;
r = 0.719,
P = 0.0010;
n = 22; Spearman's rank correlation coefficient test;
Fig. 3A). In addition, the duration of PCs was significantly correlated with the increase of vitreous temperature (
r = 0.800,
P = 0.0002;
n = 22; Spearman's rank correlation coefficient test;
Fig. 3B). On the other hand, the duration of membrane peeling in eyes with DR (14.9 ± 6.9 minutes) was not significantly correlated with the change in vitreous temperature (
r = 0.126,
P = 0.6022;
n = 18; Spearman's rank correlation coefficient test;
Fig. 3C). The correlation between the duration of core vitrectomy (3.3 ± 2.1 minutes) and the change in vitreous temperature was not significant (
r = 0.106,
P = 0.4532;
n = 51; Spearman's rank correlation coefficient test). In addition, the correlation between the duration of peripheral vitrectomy (12.3 ± 4.0 minutes) and the change in vitreous temperature was not significant (
r = 0.086,
P = 0.5995;
n = 38; Spearman's rank correlation coefficient test).
The overall mean temperature of each procedure was significantly lower in eyes that had undergone combined PEA and vitreous surgery than in eyes with vitrectomy alone (
Table 1). The temperatures before and after PC in eyes with combined cataract and vitreous surgery for DR were significantly lower than those in eyes with vitrectomy alone for DR (
P = 0.02 and 0.0070, respectively; Mann-Whitney's
U test;
Table 1). However, the number of PC in eyes undergoing combined surgery was less than that of vitrectomy alone.
The vitreous temperatures after cataract surgery in 10 patients was significantly lower than that before the surgery (P = 0.005; Wilcoxon signed-rank test). The mean vitreous temperature during PEA surgery alone was reduced by 2.3 ± 1.0°C.
The body temperature at the beginning of surgery was not significantly correlated with the change in vitreous temperatures during core vitrectomy, membrane peeling, and endophotocoagulation (r = 0.117, P = 0.3087; r = −0.142, P = 0.2253, and r = −0.067, P = 0.6376, respectively. Spearman's rank correlation coefficient test).
There were no complications such as intra- and postoperative retinal tears, retinal detachments, and endophthalmitis as a result of using a thermocouple as determined by ophthalmic fundus examinations during and after the surgery.
Our results showed that the vitreous temperature decreased significantly during core vitrectomy, peripheral vitrectomy, and cataract surgery, and increased significantly during membrane peeling and PC. There was a significant correlation between the increase in vitreous temperature and the number of PCs and the duration of the PC.
Cataract surgery decreased the vitreous temperature significantly. The temperature before PC during the combined surgery in eyes with DR was significantly lower than that in vitrectomy alone.
Electroretinographic study during vitrectomy suggested that changes in the retinal temperature can alter retinal function temporarily.
5,6 Landers et al.
4 demonstrated that the mean midvitreous temperature was 24.9°C during vitectomy when the irrigating solution was 22.4°C, which was lower than that during our vitrectomy. Our data showed that the lowest temperature in the midvitreous during surgery was 29.1 ± 1.4°C just after peripheral vitrectomy when the temperature of the irrigating solution was 25.8 ± 1.4°C.
The intraocular temperature may be dependent on the temperature of the irrigation fluid, the location of the thermocouple in the vitreous cavity, flow rate of the vitrectomy system, operating room temperature, and body temperature. It is also known that the choroid can affect the temperature in the vitreous cavity.
1–3,7 In our study, the thermocouple was inserted midway between the medial rectus muscle and inferior rectus muscle insertions, and on the opposite side of the infusion cannula in all eyes to minimize the influence of the posterior ciliary vessels. To investigate the influence of body temperature on the vitreous temperature, the correlations between the duration of core vitrectomy, membrane peeling, and endophotocoagulation and body temperature at the beginning of surgery were analyzed; however, no significant correlation was found. There is a possibility that the body temperature of each patient for each procedure might have influenced the vitreous temperature; however, the body temperature was recorded only at the beginning of surgery, and in the clinical setting during vitrectomy it might be difficult to determine the influence.
Hypothermia has been shown to protect the retina from light toxicity,
8 elevated intraocular pressures,
9 ischemic injury,
10 and to decrease postoperative inflammation in animal experiments.
10–13 Thus, it might be reasonable to decrease the vitreous temperature just before the use of bright lights such as during membrane peeling. Tamai et al.
9,14 reported that local hypothermia can protect the retina from ischemic damage induced by high intraocular pressures. They found that temperatures of 8°C and 22°C were low enough to significantly reduce the level of glutamate, a neurotoxin, in the vitreous compared to that at 38°C. However, they also showed that retinal edema developed in the inner retinal layer at both 22°C and 38°C. They concluded that a temperature of 8°C might be suitable to reduce the damage to the retina. Our results showed that the lowest temperature during the surgery was much higher than 8°C; however, no clinical adverse effects were seen by ophthalmoscopic fundus examinations and optical coherence tomography.
The temperature increased after PC. The reasons for the increase might be a slowing of the irrigation, the temperature regulation by the choroid,
1,3,7 and the transfer of heat from the choroid induced by the PC burn. In eyes with DR, the number of PCs was significantly correlated with the duration of PCs, and the duration of PCs was also significantly correlated with the change of vitreous temperature in the same group. These results suggest that stopping irrigation was the main cause of the increase of temperature during PC. On the other hand, the duration of membrane peeling was not significantly correlated with the temperature increase. Additional analysis showed that the correlation between the change in vitreous temperature and the duration of core vitrectomy or peripheral vitrectomy was not statistically significant. During vitrectomy on patients, there might be many factors that could affect the vitreous temperature during each procedure.
The average vitreous temperature during PEA decreased by 2.3 ± 1.0°C. Although the decreased temperature during vitrectomy combined with PEA might reduce the postoperative inflammation, other procedures during the course of vitrectomy might neutralize the lower temperatures. For example, PEA itself causes inflammation in the anterior chamber. Therefore, we cannot conclude that combined surgery is better than vitreous surgery alone from our findings.
The question arises as to what should be the optimal temperature of the irrigating solution to protect the retina. In patients with traumatic brain injury, therapeutic systemic hypothermia of 32.2°C to 33.9°C is used to minimize neurologic damage and reduce patient morbidity and mortality.
15,16 However, the appropriate range of temperatures is very narrow because lower temperatures (e.g., <30°C) may cause side effects, such as electrolyte imbalances and local inhibition of platelet function and clotting factors.
17 In addition, Buco et al.
18 and Honda et al.
19 reported that extremely low temperatures of 10°C in the anterior chamber can cause a breakdown of the blood–aqueous barrier. However in an ischemic model, Tamai et al.
9,14 reported that the recovery rate of a-wave and b-wave amplitudes was significantly faster, and the glutamate level in the vitreous was significantly lower at 8°C than at 38°C. Therefore, the optimal temperature in the eye remains to be determined.
Microincision vitrectomy has recently become relatively common. However, the amount of leakage of the irrigation solution through the trocar is less than that with a 20-gauge system. During membrane peeling and PC, less leakage of the solution might lead to a more rapid increase of vitreous temperature, which might not be beneficial for the light-induced toxicity and PC-related inflammation. We need to determine the optimum temperature and suitable methods to maintain it during vitrectomy.
We have not proved that the midvitreous temperature affects the retina, and hopefully future data from animal experiments will answer this question.
In conclusion, our results showed that there were consistent changes in the midvitreal temperatures during the different procedures of vitrectomy.
Supported by Grant-in-Aid for Scientific Research, the Ministry of Education, Culture, Sports, Science and Technology (no. 23390401).
Disclosure: Y. Iguchi, None; T. Asami, None; S. Ueno, None; H. Ushida, None; R. Maruko, None; K. Oiwa, None; H. Terasaki, None