Intraocular pressure (IOP) is altered by changes in arterial blood pressure, central venous pressure (CVP), intrathoracic pressure, body position, and PaCO2[1-4]. These variables are significantly altered by increased intraperitoneal pressure [5]. Moreover, anesthetics can alter IOP [6].
In a previous study of women with no preexisting eye disease, we showed that IOP was not adversely affected during gynecological laparoscopic surgery of a short duration performed under general anesthesia (GA) [7]. However, in our opinion, concern remains about a possibly deleterious effect of increased intraperitoneal pressure in patients with a preexisting narrow visual field associated with glaucoma or with increased IOP.
We designed the present study to assess the effect of pneumoperitoneum up to an intraperitoneal pressure of 15 mm Hg and of postural changes on IOP in 12 rabbits with alpha-chymotrypsin-induced glaucoma [8].
Method
The current study was conducted according to the guidelines set by our research and the animal welfare committees. Twelve male rabbits with alpha-chymotrypsin-induced intraocular hypertension were investigated.
Increased IOP was induced by injecting alpha-chymotrypsin (Iris-Pharma Laboratory, La Gaude, France) into the posterior chamber of the right eye [8]. The left eye was not injected and was used as a control. The IOP was examined daily after the alpha-chymotrypsin injection. Animals were considered for study if the sustained increase of IOP 10 mm Hg above baseline (BL) IOP persisted in the right eye for more than 6 wk. Subsequently, 5 of the 12 alpha-chymotrypsin-treated rabbits received one drop of 0.5% timolol (OphTim[registered sign]; Thea Laboratory, Clermont-Ferrand, France) in the right eye twice daily for 8 days to decrease the IOP [9]. Seven rabbits did not receive topical timolol.
Rabbits were restrained in a wooden box and were accustomed to the laboratory environment. A superficial ear vein was cannulated with a 22-gauge IV catheter through which the anesthetic was administered. Anesthesia was induced with incremental IV boluses of 18 mg of pentobarbital until lack of response to cutaneous tetanic stimulation was noted. GA was maintained with a continuous pentobarbital IV infusion at the rate of half the induction dose per hour. Animals were placed on a heating pad. A probe was placed in the rectum for continuous recording of body temperature, which was maintained between 36.8[degree sign]C and 39[degree sign]C. Subsequently, a tracheotomy was performed, a 4-mm uncuffed endotracheal tube was inserted into the trachea, and the animals were mechanically ventilated with 60% inspired oxygen in air. A single bolus of 0.1 mg/kg IV pancuronium was administered for easier initiation of mechanical ventilation after confirming adequate anesthesia depth [10]. The tidal volume was set at 7 mL/kg. The respiratory rate was adjusted throughout the study to maintain the expired CO2 between 32 and 38 mm Hg. A calibrated CO2 analyzer (Capnodig[registered sign]; Drager Medical, Antony, France) was used to measure expired CO2 from a side-sampling line placed just proximal to the endotracheal tube. Lactated Ringer's solution was administered at the rate of 5 mL [center dot] kg-1 [center dot] h-1 during the study. A femoral artery and the left external jugular vein were catheterized by surgical cutdown. The left external jugular vein was cannulated with a 20-gauge catheter. The distal portion of the catheter was advanced until the pressure tracing on the monitor indicated central positioning. An identical catheter was placed in a femoral artery to record blood pressure. These catheters were connected to a transducer positioned at the midthoracic level and referenced to atmospheric pressure. The expired CO2 curve, mean arterial pressure (MAP), heart rate (HR), and CVP were continuously recorded. For the study, the hemodynamic profile was recorded at end-expiration during a brief period of apnea. Blood gas tensions were measured once during each investigation during pneumoperitoneum with the rabbits in the head-down position.
IOP was measured by using an electronic pneumotonometer (Tono-Pen XL[registered sign]; Mentor O and O Inc., Santa Barbara, CA) after applying topical anesthesia with oxybuprocaine 0.4%. Tono-Pen XL[registered sign] is a hand-held tonometer that displays digital numeric values after contact with the cornea. Tono-Pen XL[registered sign] automatically calculates the mean IOP and the statistical reliability (standard deviation of the valid measurements) from 5% to 20% after four valid readings are obtained [11]. For this investigation, the measurement procedure was repeated until reliability of <10 percent was indicated. A single technician, well trained in the use of the instrument, made all the measurements. At each time point, the IOP of both the treated right eye (glaucomatous eye) and the untreated left eye (control eye) were measured.
Intraperitoneal insufflation of CO2 was performed during GA, with the rabbit in the supine position, at the rate of 30 mL/min, up to an intraperitoneal pressure of 15 mm Hg, via a Palmer needle placed percutaneously in the abdominal cavity. Throughout the investigation, intraperitoneal pressure was automatically maintained at 15 mm Hg by using a CO2 insufflator (Storz France, Paris, France).
Rabbits were not randomly assigned to the two experimental groups. Instead, each group (timolol and nontimolol) was investigated successively and separately. BL IOP was measured before the induction of GA, while the rabbit was conscious in the restraining box. Thereafter, MAP, CVP, HR, expired CO2, and IOP values were recorded during GA, with the rabbit in the supine position and mechanically ventilated, at the following periods: (T1) after the induction of GA, in the horizontal position, before intraperitoneal CO2 insufflation; (T2) in the horizontal position, after pneumoperitoneum had been established; (T3) pneumoperitoneum established, with a 20[degree sign] head-up tilt; (T4) pneumoperitoneum established, with a 20[degree sign] head-down tilt; (T5) pneumoperitoneum established, after return to the horizontal position; and (T6) in the horizontal position, after pneumoperitoneum had been evacuated. After each change of position and after each change of intraperitoneal pressure, a 20-min period was allowed for hemodynamic stabilization before any further measurement was taken. All measurements were performed before 2:00 PM.
At the end of the experiment, the animals were killed using an IV injection of a lethal dose of pentobarbital. A post mortem examination was performed in all animals to ensure the correct placement of the Palmer needle and to rule out any internal organ puncture.
Statistical analysis was performed separately for each group. In the timolol group, change in IOP with topical timolol compared with the pretreatment value was evaluated using Student's t-tests for paired observations. Within each group, changes in MAP, HR, and CVP over time were evaluated using a one-way analysis of variance (ANOVA) for repeated measures, followed by the Neuman-Keuls test. Within each group, changes in IOP over time were assessed using twoway ANOVA for repeated measures (the control and glaucomatous eyes were considered as two repetitions nested in each animal) [12]. Because basal values were different, analysis was based only on interaction. ANOVA was followed by the Neuman-Keuls test. Data are presented as mean +/- SD. P < 0.05 was considered the minimal value of significance.
Results
Physiological variables recorded before pneumoperitoneum are shown in Table 1. The arterial versus end-tidal CO2 difference was never more than 4 mm Hg. The HR did not change significantly in either group at any time (P = 0.9 for HR variation in both groups) (Figure 1). In nontimolol-treated rabbits, MAP differed significantly at T2 and T6 from all other time points (P = 0.01 for T2 versus all time points and T6 versus all time points), and CVP increased significantly only with pneumoperitoneum in the head-down position (P < 0.0001 for T4 versus all time points) (Figure 1). The IOP of control eyes and glaucomatous eyes did not change significantly after the induction of GA (T1) with pneumoperitoneum in the horizontal position (T2) or with pneumoperitoneum in the head-up position (T3) (Figure 1). However, the IOP increased significantly with pneumoperitoneum in the head-down position (P < 0.0001 for T4 versus all time points for glaucomatous and control eyes) (Figure 1). ANOVA showed a significant interaction between time and treatment for the IOP of control eyes and glaucomatous eyes, which demonstrates a lack of parallelism between IOP changes of control eyes and glaucomatous eyes (P = 0.024). In glaucomatous eyes, the IOP increased by 3.7 +/- 1.8 mm Hg from BL after head-down positioning (T4) (all values <5 mm Hg except one value = 7 mm Hg), whereas the IOP increased by 2.4 +/- 2.2 mm Hg (all values <5 mm Hg except one value = 6 mm Hg) from BL after head-down positioning in control eyes.
;)
Table 1: Characteristics of the Groups Before Pneumoperitoneum Had Been Created
Figure 1: Intraocular pressure (IOP), central venous pressure (CVP), mean arterial pressure (MAP), and heart rate (HR) changes in 12 rabbits with alpha-chymotrypsin-induced glaucoma in the right eye. In five rabbits, the right eye was treated with topical timolol (left panel). In seven rabbits, the right eye was not treated (right panel). The left eye was used as a control. BL = baseline, before general anesthesia, in the upright position; T1 = after induction of general anesthesia, in the horizontal position; T2 = after pneumoperitoneum (intraperitoneal pressure 15 mm Hg) had been established, in the horizontal position; T3 = during pneumoperitoneum, in the head-up position; T4 = during pneumoperitoneum in the head-down position; T5 = during pneumoperitoneum, in the horizontal position; T6 = after pneumoperitoneum had been evacuated. *P < 0.0001 versus all time points within a group.
The IOP of glaucomatous eyes decreased significantly after the topical administration of timolol compared with pretreatment values (34 +/- 5 mm Hg versus 25 +/- 5 mm Hg; P = 0.0002).
In timolol-treated rabbits, the MAP did not differ significantly among time points (P = 0.86), and the CVP increased significantly only with pneumoperitoneum in the head-down position (P < 0.0001 for T4 versus all time points) (Figure 1). The IOP of control and glaucomatous eyes decreased significantly after the induction of GA (P < 0.0001 for BL versus all time points) and increased significantly with pneumoperitoneum in the head-down position (P < 0.0001 for T4 versus all time points) (Figure 1). However, the IOP did not change significantly with pneumoperitoneum in the horizontal position (T2) or with pneumoperitoneum in the head-up position (T3) (Figure 1). For the IOP of control eyes and glaucomatous eyes, the ANOVA revealed a significant interaction between time and treatment, which demonstrates a lack of parallelism between IOP changes of control eyes and glaucomatous eyes (P < 0.001). In glaucomatous eyes, the IOP increased 3.4 +/- 1.1 mm Hg from BL after head-down positioning (T4) (all values <or=to5 mm Hg), whereas the IOP increased 2.4 +/- 0.5 mm Hg from BL after head-down positioning (T4) (all values <or=to 3 mm Hg) in control eyes.
Discussion
In a previous study of women with no preexisting eye disease undergoing gynecological laparoscopic surgery under GA, we demonstrated that intraperitoneal pressure up to 15 mm Hg does not adversely affect IOP because the decrease in IOP caused by GA compensated for the increase in IOP caused by the head-down position [7]. In the present study, the IOP of glaucomatous eyes and control eyes increased significantly only when increased intraperitoneal pressure (up to 15 mm Hg), combined with the head-down position, resulted in a significant increase in CVP. These changes occurred regardless of whether topical timolol had been administered.
The results of the present study could not be anticipated from the results of a study previously conducted in female patients without eye disease [7]. However, the results of studies evaluating the physiological and pharmacological determinants of IOP, although conducted mainly on normal eyes, support our findings. Some factors previously reported to affect IOP seemed to do so in this investigation. An instantaneous linear correlation has been demonstrated between CVP and IOP [3]. Moreover, CVP changes are accurately correlated with increased intraperitoneal pressure and with postural changes [5]. Thus, in the present investigation, in accordance with previous data [5], a CVP increase likely to result in a significant IOP increase occurred only when intraperitoneal pressure of 15 mm Hg was associated with a head-down tilt. This explains why the IOP did not change when the intraperitoneal pressure was increased to 15 mm Hg in the horizontal position and decreased slightly with an intraperitoneal pressure of 15 mm Hg in the head-up position, but increased significantly when an intraperitoneal pressure of 15 mm Hg was associated with a head-down tilt [3,5]. In addition, the response of IOP to postural changes is similarly transmitted to both glaucomatous eyes and normal eyes in humans [13]. A major finding of the present investigation is that, in timolol- and nontimolol-treated rabbits, the IOP of both glaucomatous eyes and control eyes changed similarly after pneumoperitoneum was created and after each posture change. The small studied sample may account for the statistical lack of parallelism between IOP changes of control eyes and glaucomatous eyes.
Other factors that could have altered IOP were kept constant and did not affect our results. Changes in arterial pressure within the physiological range are poorly transmitted to normal eyes [4]; however, throughout the study period, MAP remained within 10% of T1 values. IOP changes correlate linearly with expired CO2 when CVP remains unchanged [14]. However, expired CO2 was kept within a small range. Doses of pentobarbital likely to cause loss of consciousness and loss of response to painful stimuli significantly decrease IOP [15]. In this study, pentobarbital was continuously infused in all animals, and the IOP of glaucomatous eyes and control eyes was measured at each time point in the same rabbit. We did not measure intrathoracic pressure because its increase is not significantly correlated to IOP changes during laparoscopic surgery in humans [7]. Consequently, in the present investigation, neither arterial pressure, expired CO2, intrathoracic pressure changes, nor GA is likely to have altered our results.
Treated and untreated rabbits did differ in several ways. Before topical timolol was applied, the IOP of glaucomatous eyes was higher in rabbits scheduled to receive topical timolol than in those that were not. In our opinion, the nonrandomized allocation between treatment groups explains this difference in IOP. Indeed, rabbit allocation between the timolol group and the no timolol group was at the discretion of the laboratory providing the animals; thus, the groups were investigated successively. Furthermore, the IOP of the glaucomatous eye and the control eye of rabbits receiving topical timolol decreased significantly after pentobarbital administration, whereas the IOP did not change significantly after pentobarbital administration in rabbits not receiving topical timolol. Glaucoma does not explain this difference, because glaucomatous eyes and control eyes were affected similarly. Timolol pharmacological effects cannot explain this difference because IOP changes were similar in the glaucomatous eyes treated with timolol and in the control eyes not treated with timolol. Interindividual differences in anesthetic depth may have led to unintentional effects on IOP in one group. Moreover, MAP increased significantly after the establishment of pneumoperitoneum in nontimolol-treated rabbits, whereas MAP did not change significantly after the establishment of pneumoperitoneum in timolol-treated rabbits. A systemic beta-adrenergic blocking effect reported in humans after topical timolol administration [16] may explain why MAP did not increase significantly after the induction of GA in timololtreated rabbits.
Some methodological points are of interest. The arterial catheter was inserted in a lower limb artery to preserve carotid inflow, which is a determinant of IOP [4]. The central venous catheter was inserted into the left jugular vein to measure superior vena cava pressure. The left side vein was punctured to rule out any, albeit unlikely, effect on the glaucomatous eye venous return. Because of the study design, each rabbit provided both a glaucomatous eye and a healthy control eye, which were both subjected to an identical hemodynamic and pharmacological effect. The intraperitoneal pressure was increased to 15 mm Hg, which is the pressure usually achieved during laparoscopic surgery in humans [5]. Postmortem histologic examination of the alpha-chymotrypsin-treated eyes was not performed. Indeed, alpha-chymotrypsin consistently produces a chronic and stable form of glaucoma in rabbits [8,17]. The histopathologic findings in the alpha-chymotrypsin-induced glaucomatous rabbit eye correspond with those found in the human eye with long-term glaucoma [8,17]. Tono-Pen XL[registered sign] provides accurate, manometrically measured IOP values that are highly correlated to those made with the Gold-mann tonometer and to direct measurements of IOP [11]. A digital readout minimizes interobserver bias. After topical administration, timolol significantly decreases IOP in animals and in humans, whether the eyes are healthy or show ocular hypertension [9,18]. Indeed, the 40[degree sign] change of posture from a 20[degree sign] head-up to a 20[degree sign] head-down position could have caused autotransfusion of blood pooled in the venous system during the head-up position. However, a 20-minute period was allowed for hemodynamic stabilization before all IOP and CVP measurements were taken. Consequently, the CVP and IOP increases associated with a 40[degree sign] change of posture from head-up to head-down position could not have been more accentuated than they would have been after only a 20[degree sign] change of posture from the horizontal position to the head-down position. The study design minimized possible IOP alterations associated with interindividual variation in measurement technique, sex ratio, menstrual cycle, diurnal or seasonal variation, exertional changes, and refractive error [19].
Our results are potentially clinically relevant. The range of IOP recorded at BL in normal and glaucomatous eyes in the present study is consistent with the IOP values recorded in normal and glaucomatous eyes in humans [19,20]. As in humans, topical timolol alone did not restore to normal all increases in IOP, especially when BL values were very high [9,18]. After head-down positioning, the IOP increase from BL was limited to 7 mm Hg, regardless of the group. Such an IOP increase from BL is within the diurnal range of IOP variation [19] and should not to be deleterious for glaucomatous eyes, providing that IOP has been previously decreased by efficient treatment and/or that a narrow visual field did not occur [20]. Indeed, a narrow visual field resulting from increased IOP occurs when 25% of retinal cells are damaged. Therefore, the remaining functional retinal cells must be preserved [20]. Moreover, increased IOP resulting from uncontrolled glaucoma is associated with increased optic nerve weakness [20]. Consequently, sustained IOP increase during long laparoscopic procedures may worsen preexisting optic nerve damage in cases of preexisting narrow visual field and/or uncontrolled increased IOP [20]. Indeed, this animal study does not allow us to draw any definitive conclusion on the effect of increased intraperitoneal pressure on human glaucomatous eyes. Nevertheless, we suggest that laparoscopic surgery should be discouraged in patients with preoperative uncontrolled increased IOP and/or narrow visual field until it is demonstrated in human glaucomatous eyes that the decrease of IOP caused by GA compensates for the IOP increase caused by head-down positioning at increased abdominal pressures, as it does in normal eyes [7].
In conclusion, the effect of increased intraperitoneal pressure (up to 15 mm Hg) created by CO2 insufflation was investigated in glaucomatous eyes during procedures of two hours' duration. The major finding of this study is that, in a reliable model of glaucoma, CO2 pneumoperitoneum was associated with a clinically relevant increase in IOP when a head-down position was combined with pneumoperitoneum. Although this is an animal study, we believe that these data, together with those previously obtained in humans with healthy eyes [5], suggest that laparoscopic surgery is not harmful to patients with preexisting glaucoma, providing that visual fields are not narrow and/or that IOP is adequately controlled before surgery.
The authors gratefully acknowledge the valuable advice of Dr. Francoise Niessen, staff ophthalmologist.
REFERENCES
1. Cunningham AJ, Barry P. Intraocular pressure: physiology and implications for anaesthetic management. Can J Anaesth Soc 1986;33:198-208.
2. Libonati MM. General anesthesia. In: Tagman W, Jaeger EA, eds. Duane's clinical ophthalmology. Vol 6. Philadelphia: JB Lippincott, 1992:1-12.
3. Macri FJ. Interdependence of venous and eye pressure. Arch Ophthalmol 1961;65:150-7.
4. Riva CE, Sinclair SH, Grunwald JE. Autoregulation of retinal circulation in response to decrease of perfusion pressure. Invest Ophthalmol Vis Sci 1981;21:34-8.
5. Gannedahl P, Odeberg S, Brodin L-A, Sollevi A. Effects of posture and pneumoperitoneum during anaesthesia on the indices of left ventricular filling. Acta Anaesthesiol Scand 1996;40:160-6.
6. Murphy DF. Anesthesia and intraocular pressure. Anesth Analg 1985;64:520-30.
7. Lentschener C, Benhamou D, Niessen F, et al. Intraocular pressure changes during gynaecological laparoscopy. Anaesthesia 1996;51:1006-8.
8. Chee P, Hamasaki DI, Fla M. The basis for chymotrypsin-induced glaucoma. Arch Ophthalmol 1971;85:103-6.
9. Heel RC, Brogden RN, Speight TM, Avery GS. Timolol: a review of its therapeutic efficacy in the topical treatment of glaucoma. Drugs 1979;17:38-55.
10. Drummond JC, Todd MM, Saidman LJ. Use of neuromuscular blocking drugs in scientific investigations involving animal subjects. Anesthesiology 1996;85:697-9.
11. Minckler DS, Baerveldt G, Heuer DK, et al. Clinical evaluation of the oculab Tono-Pen. Am J Ophthalmol 1987;104:168-73.
12. Winner BL. Statistical principles in experimental design. 2nd ed. New York: McGraw Hill, 1971:539-75.
13. Krieglstein GK, Langham ME. Influence of body position on the intraocular pressure of normal and glaucomatous eyes. Ophthalmol Basel 1975;171:132-45.
14. Hvidberg A, Kessing SVV, Fernandez A. Effect of changes in PCO
2 and body positions on intraocular pressure during general anaesthesia. Acta Ophthalmol (Copenh) 1981;59:465-75.
15. Donlon JV Jr. Anesthesia and eye, ear, nose and throat surgery. In: Miller RD, ed. Anesthesia. 4th ed. New York: Churchill Livingstone, 1994:175-96.
16. Backlund M, Kirvela M, Lindgren L. Cardiac failure aggravated by timolol eye drops: preoperative improvement by changing to pilocarpine. Acta Anaesthesiol Scand 1996;40:379-81.
17. Best M, Rabinovitz AZ, Masket S. Experimental alpha-chymotrypsin glaucoma. Ann Ophthalmol 1975;7:803-10.
18. Vareilles P, Silverstone D, Plazonnet B, et al. Comparison of the effects of timolol and other adrenergic agents on intraocular pressure in the rabbit. Invest Ophthalmol Vis Sci 1977;16:987-96.
19. Shiose Y. Intraocular pressure: new perspectives. Surv Ophthalmol 1990;34:413-35.
20. Scheie HG, Albert DM. Glaucoma, textbook of ophthalmology. 9th ed. Philadelphia: WB Saunders, 1977:522-53.
