- Email: [email protected]
Safety Profile of Venturi Versus Peristaltic Phacoemulsification Pumps in Cataract Surgery Using a Capsular Surrogate for the Human Lens
Safety Profile of Venturi Versus Peristaltic Phacoemulsification Pumps in Cataract Surgery Using a Capsular Surrogate for the Human Lens MICHAEL GILBERT, BRIAN ZAUGG, BRIAN STAGG, AND RANDALL J. OLSON PURPOSE:
To compare the risk of capsular rupture of the human lens during cataract surgery from contact by phacoemulsification needles using different vacuum pumps, ultrasound modalities, and contact angles. DESIGN: Experimental laboratory investigation. METHODS: The John A. Moran Eye Center, University of Utah, Salt Lake City, Utah, was the setting for this study. A Signature (Abbott Medical Optics, Inc) phacoemulsification machine was used in peristaltic and Venturi vacuum modes with transversal and micropulsed ultrasound. Contact was made with a capsular surrogate to achieve tip occlusion or tip contact only. Breakage rates were calculated by analyzing the capsular surrogate under a surgical microscope. RESULTS: Venturi and peristaltic pump modes had similar risk of capsular rupture, regardless of whether the data were analyzed with tip occlusion data included (44.2% peristaltic vs 40.2% Venturi, P [ .047) or excluded from the analysis (66.3% peristaltic vs 60.3% Venturi, P [ .013). Transversal ultrasound was significantly more likely to cause capsular rupture than micropulsed ultrasound (69.8% vs 56.8%, P < .0001). Tip contact was significantly more likely than tip occlusion to cause capsular rupture (63.3% vs 0%, P < .0001). CONCLUSIONS: There is no significant difference in risk of capsular rupture using Venturi rather than peristaltic vacuum pumps, while transversal seemed to increase the risk when compared to micropulsed ultrasound. Tip occlusion is not a risk factor for capsular rupture, as all breaks in the capsular surrogate occurred with tip contact. (Am J Ophthalmol 2015;160(1): 179–184. Ó 2015 by Elsevier Inc. All rights reserved.)
M
ANY ADVANCES HAVE BEEN MADE IN CATARACT
surgery since its advent, especially since the addition of phacoemulsification by Dr Charles Kelman.1 However, with the addition of this powerful
Accepted for publication Apr 10, 2015. From the Department of Ophthalmology and Visual Sciences, John A. Moran Eye Center, University of Utah, Salt Lake City, Utah. Mr. Gilbert is now at Creighton University School of Medicine, Omaha, Nebraska. Inquiries to Randall J. Olson, John A. Moran Eye Center, University of Utah, 65 Mario Capecchi Drive, Salt Lake City, UT 84132; e-mail: [email protected] 0002-9394/$36.00 http://dx.doi.org/10.1016/j.ajo.2015.04.017
Ó
2015 BY
technology came additional opportunities for complications. Posterior capsular rupture is a relatively common complication of cataract extraction via phacoemulsification, with a reported incidence of between 1.7% and 5.2%.2–6 Capsular rupture can occur during many different steps of cataract surgery but most frequently occurs during nucleus removal,5 with approximately 40%–60% of capsular ruptures occurring during this stage.5,7,8 Most likely, this complication is owing to the combination of ultrasound energy and the sharp needle tip. While outcomes after capsular rupture are typically good,2–4,6–8 there is a definite increase in the incidence of temporarily raised intraocular pressure, persistent uveitis, endophthalmitis, cystoid macular edema, retinal detachment, and retained nuclear material, complications that in some cases necessitate additional surgery.2,3,7,8 Thus, many advances in phacoemulsification technique and technology have centered on reducing the risk of capsular rupture to increase the safety of the procedure. Given the wide array of options available to the cataract surgeon regarding types of vacuum pumps, needle specifications, and ultrasound modalities, little data exist comparing the safety profiles of these different options. Recent studies have evaluated optimization of phacoemulsification variables for maximum efficiency of nuclear emulsification and decreased chatter9–12; however, all of those optimized configurations were found with the machines in peristaltic mode. A recent study by our group showed that Venturi vacuum significantly improved the efficiency of lens fragment removal when compared to peristaltic vacuum.13 Owing to the inherent differences in vacuum generation and fluid dynamics between peristaltic and Venturi pumps, their safety profiles cannot be assumed to be identical. This study assesses the risk of rupture if the phacoemulsification needle were to inadvertently contact the capsule during cataract surgery, and compares optimized phacoemulsification settings in peristaltic and Venturi mode.
METHODS DESIGN:
This study was designed as an experimental laboratory investigation. No human or animal studies were done as part of this investigation.
ELSEVIER INC. ALL
RIGHTS RESERVED.
179
PREVIOUS WORK:
Our methods and experimental design are a continuation and expansion of work by Meyer and associates.14 This previous study used peristaltic modes on the Infiniti (Alcon, Inc, Fort Worth, Texas, USA) and Signature (Abbot Medical Optics [AMO], Inc, Santa Ana, California, USA) phacoemulsification machines with a bottle height of 75 cm, machine-indicated flow rate of 60 mL/min, and 550 mm Hg vacuum setting. Their micropulsed ultrasound trials used a WhiteStar handpiece (AMO, Inc) at 2 longitudinal power settings, 10% and 100%, both with a 6 ms on-duty cycle and 12 ms off-duty cycle. Their transversal ultrasound trials used an Ellips handpiece (AMO, Inc) at 100% power and torsional ultrasound trials used an OZil handpiece (Alcon, Inc) at 100% power. For each of these 4 ultrasound modalities, they tested 4 needles: (1) 19 gauge sharp, (2) 19 gauge Dewey Radius (MST, Redmond, Washington, USA), (3) 20 gauge sharp, and (4) 20 gauge Dewey Radius. For each condition, the tip was tapped against the capsule surrogate 200 times. Additionally, they tested 20 fresh human cadaver lenses with exactly the same phacoemulsification parameters. However, owing to the limited number of cadaver lenses, they were limited to the following conditions: (1) micropulsed ultrasound at 100% power with a 6 ms on-duty cycle and 12 ms off-duty cycle with a 20 gauge Dewey Radius needle, (2) transversal ultrasound at 100% power with a 20 gauge sharp needle, (3) transversal ultrasound at 100% power with a 20 gauge Dewey Radius needle, and (4) torsional ultrasound at 100% power with a 20 gauge Dewey Radius needle. The lenses were tapped gently in different places until capsule rupture occurred, rendering the lenses unusable for further trials.14 They found the rounded edge tip to be very protective of the capsule with all modalities tested.
CAPSULE SURROGATE:
Plastic wrap (Great Value Clear Plastic Wrap; Wal-Mart Stores, Inc, Bentonville, Arkansas, USA) was stretched tightly over one end of a closed 4-inch-diameter polyvinylchloride tube as a surrogate for the capsule in the manner previously described by Meyer and associates.14 The Meyer study used fresh whole human cadaver lenses to further validate this approach. All trials were performed on plastic wrap from the same roll, and pieces of plastic wrap were cut to be large enough to extend beyond the edge of the tube by at least 1 inch in all directions. Tension on the capsule surrogate was established by placing it over the pipe underwater to ensure no air bubbles were trapped underneath, then placing a rubber band around the diameter of the tube to secure it. Once secured in this fashion, to ensure a reasonably uniform tension in all directions, the capsule surrogate was examined for wrinkling, as this would indicate uneven tension. Any wrinkling was resolved by adjusting tension in the appropriate axis of the plastic wrap. This adjustment process was repeated until the entire surface of the capsule surrogate was smoothly uniform. This formed a chamber that was 180
completely filled with water and then submerged in a water bath in order to have a fluid interface on both sides of the capsule surrogate. PHACOEMULSIFICATION
Signature SETTINGS: The (AMO, Inc) phacoemulsification machine was used with the Fusion cassette (AMO, Inc), with all trials using a 550 mm Hg vacuum setting in both peristaltic and Venturi modes, a bottle height of 50 cm, and balanced salt solution. All trials in peristaltic mode used a flow rate of 40 mL/min. While we do not know the exact fluid flow for Venturi vacuum at 550 mm Hg, in a previous study at 500 mm Hg fluid flow was 102.2 6 2.3 mL/min for micropulsed longitudinal and 96.6 6 0.7 mL/min for transversal ultrasound.13 Micropulsed ultrasound trials used a WhiteStar handpiece (AMO, Inc), while transversal ultrasound trials used an Ellips FX handpiece (AMO, Inc). Micropulsed ultrasound was set at 50% power with a 6 ms on-duty cycle and 6 ms off-cycle.15,16 Transversal ultrasound was set at continuous 50% power. Ultrasound and vacuum were both used at only their maximum setting, with all settings on panel control to ensure uniformity. While previous work9 has shown that micropulsed ultrasound is most efficient at 100% power and transversal at 50% power, we decided the fairest test would be to use the same power setting for both in order to minimize power settings as a confounding factor if there were differing breakage rates.
PHACOEMULSIFICATION TIPS:
All trials used phacoemulsification tips with a 30 degree bevel and 0.9 mm (20 gauge) tip diameter. Transversal ultrasound trials were done with a 30 degree bent (Kelman) tip, while micropulsed trials used a straight tip. All tips were made by MicroSurgical Technology (Redmond, Washington, USA). A new tip was used at the beginning of each condition to control for the effect of repeated contact with the capsule surrogate on tip sharpness.
CONTACT WITH CAPSULE SURROGATE:
In order to distinguish possible differences in safety between peristaltic and Venturi pump modes, 2 techniques for contacting the capsule surrogate simulated different clinical situations. The first technique consisted of tapping the edge of the tip against the plastic wrap by approaching the membrane with the needle bevel up at an angle approximately 30 degrees above parallel to the testing membrane, with no tip occlusion, as indicated in the Figure, Left (tip contact). Ultrasound was engaged during the approach to the surrogate capsule, needle contact, and retraction away from the membrane. The second technique involved tapping the needle opening at an appropriate angle against the plastic wrap to achieve tip occlusion. To achieve proper and even occlusion, the handle was held such that the bevel of the needle was facing downward with the needle opening parallel to the capsule surrogate as indicated in the Figure,
AMERICAN JOURNAL OF OPHTHALMOLOGY
JULY 2015
FIGURE. Two approaches were used to contact a surrogate of the human lens capsule with a phacoemulsification needle, in order to assess the risk of capsule rupture using peristaltic and Venturi vacuum pumps with micropulsed and transversal ultrasound. The line represents the capsule surrogate. (Left) Tip contact: The needle is held bevel up, at an angle approach approximately 30 degrees from the capsule surrogate. (Right) Tip occlusion: The needle is held bevel down, with the needle bevel opening parallel to the capsule surrogate.
Right (tip occlusion). Ultrasound was engaged during approach to the surrogate and contact and establishment of occlusion; once tip occlusion was indicated by the increasing pitch of audio tones, which signaled increasing vacuum, the pedal was released fully and the tip was simultaneously retracted away from the plastic wrap. The total time of occlusion from initial contact of the needle to release and retraction away from the membrane was approximately 1 second. In both techniques, contact between the needle and capsule surrogate was monitored under direct visualization, which was aided by placement of a light that would reflect off the membrane, allowing observation of the time and point of contact. Eight conditions were tested: (1) micropulsed ultrasound in peristaltic mode with tip contact; (2) micropulsed ultrasound in peristaltic mode with tip occlusion; (3) transversal ultrasound in peristaltic mode with tip contact; (4) transversal ultrasound in peristaltic mode with tip occlusion; (5) micropulsed ultrasound in Venturi mode with tip contact; (6) micropulsed ultrasound in Venturi mode with tip occlusion; (7) transversal ultrasound in Venturi mode with tip contact; and (8) transversal ultrasound in Venturi mode with tip occlusion. For each occlusion condition, contact was made with the plastic wrap in different places for a total of 200 taps. For each tip contact, contact was made with the plastic wrap in different places for a total of 400 taps. This differential in the number of taps was owing to the considerable difference in breakage rates between tip occlusion and tip contact conditions that we observed in preliminary experiments. ANALYSIS OF CAPSULE SURROGATE:
After each run of 200 or 400 taps, the plastic wrap was examined under a microsurgical microscope (Leica Microsystems Inc, Buffalo Grove, Illinois, USA) using a Sinskey hook (Bausch & Lomb Inc, Rochester, New York, USA) to gently probe
VOL. 160, NO. 1
for breakage. Additionally, in tip occlusion trials we further verified that occlusion had been achieved by observing a full 360 degree outline of the needle in the plastic wrap for each trial. A x2 analysis was used to compare breakage rates between conditions, looking at differences between peristaltic and Venturi modes; between micropulsed and transversal ultrasound; and between tip edge contact and tip occlusion conditions. Results were considered statistically significant at P < .007 after a Bonferroni correction for multiple comparisons.
STATISTICAL ANALYSIS:
RESULTS SEVEN COMPARISONS WERE PERFORMED ON THE DATA SET:
(1) peristaltic vs Venturi pump modes with occlusion data included, (2) peristaltic vs Venturi pump modes without occlusion data, (3) micropulsed vs transversal ultrasound with occlusion data included, (4) micropulsed vs transversal ultrasound without occlusion data, (5) tip contact vs tip occlusion techniques, (6) tip contact using micropulsed ultrasound in peristaltic vs Venturi pump modes, and (7) tip contact using transversal ultrasound in peristaltic vs Venturi pump modes. We analyzed the peristaltic vs Venturi pump data and the micropulsed vs transversal ultrasound data twice, including the occlusion data in one comparison and excluding it in another. This is because with the occlusion techniques, in all 800 trials and regardless of ultrasound modality or vacuum pump type, not a single break occurred. Hence, in comparisons where the tip occlusion data were included, the large number of included trials that did not cause any breaks resulted in a smaller percentage difference
SAFETY OF VENTURI VERSUS PERISTALTIC PHACOEMULSIFICATION PUMPS
181
TABLE. Breaks in a Capsular Surrogate for the Human Lens for Peristaltic and Venturi Phacoemulsification Pumps, Using Micropulsed and Transversal Ultrasound Modalities, and Tip Contact or Tip Occlusion Techniques, With Machine Set at 550 mm Hg, 50 cm Bottle Height (Actual), and When in Peristaltic Mode a 40 mL/min Flow Rate (Machine Indicated) Conditions
Vacuum Pump Type
Ultrasound Modality
Tip Contact or Occlusion
1 2 3 4 5 6 7 8
Peristaltic Peristaltic Peristaltic Peristaltic Venturi Venturi Venturi Venturi
Micropulsed Micropulsed Transversal Transversal Micropulsed Micropulsed Transversal Transversal
Contact Occlusion Contact Occlusion Contact Occlusion Contact Occlusion
No. of Trials Performed
No. (%) of Breaks
400 200 400 200 400 200 400 200
246 (61.5%) 0 284 (71%) 0 208 (52%) 0 274 (68.5%) 0
P < .0001 for all tip contact vs occlusion (rows 1, 3, 5, & 7 vs 2, 4, 6, & 8). P ¼ .013 peristaltic vs Venturi for all comparisons with contact (rows 1 & 3 vs 5 & 7). P < .0001 for all micropulsed vs transversal with contact (rows 1 & 5 vs 3 & 7). P ¼ .0067 for micropulsed contact only peristaltic vs Venturi (row 1 vs 5). P ¼ .44 for transversal contact only peristaltic vs Venturi (row 3 vs 7).
in breakage rates. Thus, in an effort to unmask possible smaller differences between variables, we analyzed the data both ways. We found that there was no significant difference in breakage rates between peristaltic and Venturi pump modes. When tip occlusion data were included in this comparison, breakage rates were 44.2% for peristaltic pump mode and 40.2% for Venturi pump mode (P ¼ .047). When tip occlusion data were excluded from this comparison, breakage rates were 66.3% for peristaltic pump mode and 60.3% for Venturi pump mode (P ¼ .013). Neither of these comparisons exceeded our corrected level of statistical significance. We further found that transversal ultrasound was significantly more likely to cause a breakage than micropulsed ultrasound. When tip occlusion data were included in this comparison, breakage rates were 46.5% for transversal ultrasound and 37.8% for micropulsed ultrasound (P < .0001). When tip occlusion data were excluded from this comparison, breakage rates were 69.8% for transversal ultrasound and 56.8% for micropulsed ultrasound (P < .0001). Both these comparisons exceeded our corrected level of statistical significance. As stated above, not a single break occurred in any of our tip occlusion trials, regardless of ultrasound modality or vacuum pump type. Thus, contacting the capsular surrogate with the tip was significantly more likely to cause a breakage than tip occlusion, with a breakage rate of 63.3% for tip contact and 0% for tip occlusion (P < .0001) (Table). In an attempt to isolate the effect of switching between peristaltic and Venturi vacuum pump modes, an additional 2 comparisons were identified and analyzed. First, breakage rates with tip contact using micropulsed ultrasound were 61.5% for peristaltic pump mode and 52% using Venturi 182
pump mode, which reached statistical significance (P ¼ .0067). Second, breakage rates with tip contact using transversal ultrasound were 71% for peristaltic pump mode and 68.5% for Venturi pump mode, which was not significant (P ¼ .4415).
DISCUSSION CAPSULE RUPTURE MOST COMMONLY OCCURS DURING THE
phacoemulsification stage of cataract surgery. Although surgeons make every attempt to avoid contacting the capsule with the phacoemulsification needle, occasional inadvertent contact is inevitable. Meyer and associates14 observed that 6 variables have an effect on likelihood of capsule breakage: (1) the amount of pressure of the tip against the capsule; (2) the amount of active vacuum at the tip; (3) the speed of flow of the machine; (4) the gauge of the needle; (5) the needle sharpness and degree of angulation; and (6) the energy modulation active at the time of contact. Our study examined the second of these variables, which results from the inherent differences between peristaltic and Venturi pumps in fluidics at the phacoemulsification tip. The results of this study suggest 2 additional variables: (7) the ultrasound modality in use; and (8) the angle at which the tip contacts the capsule. Many surgeons have observed both anecdotally and experimentally that use of Venturi pumps can decrease surgical time, but at the cost of higher postocclusion surge and shorter time to react to events that occur during phacoemulsification and aspiration of cataractous nuclear material.17–20 Contrary to expectations and paradoxically, our results show that the Venturi pump may have a marginally lower risk of capsular breakage than the
AMERICAN JOURNAL OF OPHTHALMOLOGY
JULY 2015
peristaltic pump, although this difference did not quite reach statistical significance (P ¼ .0128). Once we included tip occlusion data in the comparison, the difference between breakage rates for peristaltic and Venturi systems clearly was not significant. One possible explanation is that unoccluded flow vacuum in peristaltic systems is greater than some might assume.14,18,21,22 This would decrease the anticipated difference between the 2 systems in vacuum produced at the phacoemulsification tip, even without occlusion. It is difficult to explain why peristaltic vacuum might be worse. However, we believe our results support the conclusion that at the very least, there is no significant difference between Venturi and peristaltic pumps in risk of capsular breakage. Our analysis revealed 2 other notable findings, both of which achieved statistical significance. The first was the difference in breakage rates between transversal and micropulsed ultrasound modalities, with significantly higher breakage rates in the transversal ultrasound trials. This finding persisted regardless of whether the tip occlusion data were included or excluded. In this case, it may be that the transversal motion produces more of a sideways ‘‘slicing’’ effect on the membrane, while the traditional micropulsed ‘‘jackhammer’’ motion simply impacts and stretches the membrane with less likelihood of rupturing it. Secondly, we found that contact of the tip edge with the membrane was significantly more likely to cause breakage than with full occlusion of the tip by the membrane. Indeed, during a total of 800 trials with both ultrasound modalities and both vacuum pump types in which we produced occlusion, not a single break occurred. Given the relatively high vacuum settings used in this experiment, this result is surprising, particularly because of the theoretically higher unoccluded flow vacuum that the Venturi pump creates. This finding, in conjunction with our initial observation that the Venturi and peristaltic modes likely had fairly similar breakage rates, suggests that the main potential for phacoemulsification-associated capsular rupture may be a consequence of the inherent sharpness of the needle. The Meyer study, which documented that a radiused edge on the phacoemulsification needle, where all sharpness was removed, dramatically reduced the risk of capsular breakage, provides further support of this possibility. Thus, it is also possible that the magnitude of this effect may differ based on the variability in tip finishes from different needle manufacturers or differences in the edge over time, especially small barbs. More investigation, especially of torsional motion, is needed to further explore these findings. Our isolated peristaltic pump vs Venturi pump comparison revealed an interesting contrast between micropulsed and transversal ultrasound. Namely, when tip occlusion data were excluded and only tip contact data were considered, with micropulsed ultrasound there was a statistically significant difference in breakage rates, while with transVOL. 160, NO. 1
versal ultrasound the difference in breakage rates was not significant. One possible explanation for this difference rests with the different movements present at the needle tip in each ultrasound modality. As we conjectured above, the higher breakage rates seen in transversal ultrasound may be attributed to a sideways ‘‘slicing’’ effect not present in the micropulsed ultrasound modality. It may be that in the presence of a larger risk factor for capsule rupture, in this case transversal ultrasound, the difference made in breakage rates by the vacuum pump type becomes less of a factor. Contrast this with micropulsed ultrasound, in which the ‘‘jackhammer’’ motion lowers the risk of capsule rupture but unmasks the more subtle risk of differing vacuum pumps. As with our prior discussion above, it is difficult to explain why the peristaltic pump might be worse; however, we should note that this finding only exceeded our significance cutoff by a very small margin and thus should not be given the same weight as our other findings. Additionally, these comparisons were done excluding tip occlusion data, magnifying differences between other variables. We again emphasize that at the very least our results clearly support the conclusion that there is no significant difference in the risk of capsular breakage between Venturi and peristaltic pumps. From a clinical perspective, several important conclusions can be drawn from our results and put into practice. Most clearly demonstrated is the difference in breakage rates between tip contact and tip occlusion. In surgical practice, this most closely mirrors the ‘‘bevel-up’’ and ‘‘bevel-down’’ approaches. Contrary to common belief, our data support a bevel-down approach to phacoemulsification from the standpoint of reducing the risk of capsule rupture. This is because with a bevel-down approach, in the event of inadvertent contact with the capsule the surgeon would increase the likelihood of a tip occlusion–type event and decrease the likelihood of a tip contact–type event. Furthermore, we demonstrated that using a Venturi vacuum pump in cataract surgery does not increase the risk of capsule rupture. While our data do not address some of the common comparisons of Venturi systems having higher postocclusion surge and shorter reaction time intraoperatively compared to peristaltic systems, they do support the conclusion that Venturi and peristaltic vacuum pumps can be considered equivalent in their risk of capsule rupture. Thus, surgeons should consider other variables, such as efficiency, effective phacoemulsification time, and personal preference, when evaluating vacuum pump types. The principal limitation of this study is that our methods are simulations of situations that can occur during cataract surgery. It would not be possible to conduct such an experiment using human subjects. Additionally, a study with methods similar to ours14 used a relatively small sample of human cadaver capsules in addition to the capsular surrogate; their results from the cadaver capsules were consistent with what they observed using plastic wrap as a
SAFETY OF VENTURI VERSUS PERISTALTIC PHACOEMULSIFICATION PUMPS
183
capsular surrogate. Our results would be prohibitively expensive to investigate and replicate using human cadaver capsules, and we believe that our methods and procedures provide a feasible and effective way to accomplish our objectives. At the very least the risk ratio is likely to be similar, even if the absolute risk is likely to be somewhat different. In conclusion, it is important to note that the risk of capsule rupture is directly related to direct contact with
the capsule by the sharp point of the tip and is not secondary to occlusion even with ultrasound on. Because of this finding, we did not see a significant difference in capsular contact risk between Venturi and peristaltic vacuum even when we used very aggressive parameters. The use of transversal ultrasound appears to increase capsular rupture risk, and our previous work has shown that edge sharpness of the tip is an additional risk factor. Other tip motions need to be assessed in an equally rigorous fashion.
ALL AUTHORS HAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST. Mr Gilbert, Dr Zaugg, Dr Stagg, and Dr Olson do not have any proprietary or financial interest, or any other conflicts of interest, to report. This study was supported in part by an unrestricted grant from Research to Prevent Blindness, Inc, New York, New York, USA, to the Department of Ophthalmology and Visual Sciences, University of Utah, Salt Lake City, Utah, USA. Dr Stagg and Dr Zaugg are recipients of Achievement Rewards for College Scientists Foundation, Inc (ARCS), Utah Chapter, Salt Lake City, Utah, USA, Scholar Awards. MicroSurgical Technology (Redmond, Washington, USA) donated the tips that were used for the study. All authors attest that they meet the current ICMJE requirements to qualify as authors. Susan Schulman, University of Utah School of Medicine, Salt Lake City, Utah, served as a consulting medical writer.
REFERENCES 1. Pandey SK, Milverton EJ, Maloof AJ. A tribute to Charles David Kelman MD: ophthalmologist, inventor and pioneer of phacoemulsification surgery. Clin Experiment Ophthalmol 2004;32(5):529–533. 2. Ang GS, Whyte IF. Effect and outcomes of posterior capsule rupture in a district general hospital setting. J Cataract Refract Surg 2006;32(4):623–627. 3. Chan FM, Mathur R, Ku JJ, et al. Short-term outcomes in eyes with posterior capsule rupture during cataract surgery. J Cataract Refract Surg 2003;29(3):537–541. 4. Ionides A, Minassian D, Tuft S. Visual outcome following posterior capsule rupture during cataract surgery. Br J Ophthalmol 2001;85(2):222–224. 5. Pingree MF, Crandall AS, Olson RJ. Cataract surgery complications in 1 year at an academic institution. J Cataract Refract Surg 1999;25(5):705–708. 6. Wilczy nski M, Wilczy nska O, Synder A, Omulecki W. Incidence and functional outcome of phacoemulsification complicated by posterior capsular rupture. Klin Oczna 2009; 111:26–29. 7. Ti SE, Yang YN, Lang SS, Chee SP. A 5-year audit of cataract surgery outcomes after posterior capsule rupture and risk factors affecting visual acuity. Am J Ophthalmol 2014;157(1): 180–185. 8. Osher RH, Cionni RJ. The torn posterior capsule: its intraoperative behavior, surgical management, and long-term consequences. J Cataract Refract Surg 1990;16(4):490–494. 9. DeMill DL, Zaugg BE, Pettey JH, et al. Objective comparison of 4 nonlongitudinal ultrasound modalities regarding efficiency and chatter. J Cataract Refract Surg 2012;38(6):1065–1071. 10. Oakey ZB, Jensen JD, Zaugg BE, Radmall BR, Pettey JH, Olson RJ. Porcine lens nuclei as a model for comparison of 3 ultrasound modalities regarding efficiency and chatter. J Cataract Refract Surg 2013;39(8):1248–1253. 11. Farukhi AM, Stagg BC, Ronquillo C Jr, et al. Effect of phaco tip diameter on efficiency and chatter. J Cataract Refract Surg 2014;40(5):811–817.
184
12. Gupta I, Zaugg B, Stagg BC, et al. Phacoemulsification efficiency with a radiused phaco tip. J Cataract Refract Surg 2014;40(5):818–821. 13. Cahoon JM, Gupta I, Gardiner G, et al. A comparison between Venturi and peristaltic vacuum in phacoemulsification. J Cataract Refract Surg 2015;41(2):428–432. 14. Meyer JJ, Kuo AF, Olson RJ. The risk of capsular breakage from phacoemulsification needle contact with the lens capsule: a laboratory study. Am J Ophthalmol 2010;149(6): 882–886. 15. Kirk KR, Ronquillo C Jr, Jensen JD, et al. Optimum on-time duty cycle for micropulse technology. J Cataract Refract Surg 2014;40(9):1545–1548. 16. Jensen JD, Kirk KR, Gupta I, et al. Determining optimal ultrasound off time with micropulsed longitudinal phacoemulsification. J Cataract Refract Surg 2015;41(2):433–436. 17. Payne M, Georgescu D, Waite AN, Olson RJ. Phacoemulsification tip vacuum pressure: comparison of 4 devices. J Cataract Refract Surg 2006;32(8):1374–1377. 18. Floyd MS, Valentine JR, Olson RJ. Fluidics and heat generation of Alcon Infiniti and Legacy, Bausch & Lomb Millennium, and Advanced Medical Optics Sovereign phacoemulsification systems. Am J Ophthalmol 2006;142(3): 387–392. 19. Georgescu D, Payne M, Olson RJ. Objective measurement of postocclusion surge during phacoemulsification in human eye-bank eyes. Am J Ophthalmol 2007;143(3):437–440. 20. Patricio MS, Almeida AC, Rodrigues MP, Guedes ME, Ferreira TB. Correlation between cataract grading by Scheimpflug imaging and phaco time in phacoemulsification using peristaltic and Venturi pumps. Eur J Ophthalmol 2013; 23(6):789–792. 21. Georgescu D, Kuo AF, Kinard KI, Olson RJ. A fluidics comparison of Alcon Infiniti, Bausch & Lomb Stellaris, and Advanced Medical Optics Signature phacoemulsification machines. Am J Ophthalmol 2008;145(6):1014–1017. 22. Brinton JP, Adams W, Kumar R, Olson RJ. Comparison of thermal features associated with 2 phacoemulsification machines. J Cataract Refract Surg 2006;32(2):288–293.
AMERICAN JOURNAL OF OPHTHALMOLOGY
JULY 2015
Biosketch Michael Gilbert, MD, graduated from the Creighton University School of Medicine. He is currently an intern at the University of Kansas Medical Center and will be continuing there as a resident in the Department of Ophthalmology in 2016.
VOL. 160, NO. 1
SAFETY OF VENTURI VERSUS PERISTALTIC PHACOEMULSIFICATION PUMPS
184.e1
©2015 Elsevier
To compare the risk of capsular rupture of the human lens during cataract surgery from contact by phacoemulsification needles using different vacuum pumps, ultrasound modalities, and contact angles. DESIGN: Experimental laboratory investigation. METHODS: The John A. Moran Eye Center, University of Utah, Salt Lake City, Utah, was the setting for this study. A Signature (Abbott Medical Optics, Inc) phacoemulsification machine was used in peristaltic and Venturi vacuum modes with transversal and micropulsed ultrasound. Contact was made with a capsular surrogate to achieve tip occlusion or tip contact only. Breakage rates were calculated by analyzing the capsular surrogate under a surgical microscope. RESULTS: Venturi and peristaltic pump modes had similar risk of capsular rupture, regardless of whether the data were analyzed with tip occlusion data included (44.2% peristaltic vs 40.2% Venturi, P [ .047) or excluded from the analysis (66.3% peristaltic vs 60.3% Venturi, P [ .013). Transversal ultrasound was significantly more likely to cause capsular rupture than micropulsed ultrasound (69.8% vs 56.8%, P < .0001). Tip contact was significantly more likely than tip occlusion to cause capsular rupture (63.3% vs 0%, P < .0001). CONCLUSIONS: There is no significant difference in risk of capsular rupture using Venturi rather than peristaltic vacuum pumps, while transversal seemed to increase the risk when compared to micropulsed ultrasound. Tip occlusion is not a risk factor for capsular rupture, as all breaks in the capsular surrogate occurred with tip contact. (Am J Ophthalmol 2015;160(1): 179–184. Ó 2015 by Elsevier Inc. All rights reserved.)
M
ANY ADVANCES HAVE BEEN MADE IN CATARACT
surgery since its advent, especially since the addition of phacoemulsification by Dr Charles Kelman.1 However, with the addition of this powerful
Accepted for publication Apr 10, 2015. From the Department of Ophthalmology and Visual Sciences, John A. Moran Eye Center, University of Utah, Salt Lake City, Utah. Mr. Gilbert is now at Creighton University School of Medicine, Omaha, Nebraska. Inquiries to Randall J. Olson, John A. Moran Eye Center, University of Utah, 65 Mario Capecchi Drive, Salt Lake City, UT 84132; e-mail: [email protected] 0002-9394/$36.00 http://dx.doi.org/10.1016/j.ajo.2015.04.017
Ó
2015 BY
technology came additional opportunities for complications. Posterior capsular rupture is a relatively common complication of cataract extraction via phacoemulsification, with a reported incidence of between 1.7% and 5.2%.2–6 Capsular rupture can occur during many different steps of cataract surgery but most frequently occurs during nucleus removal,5 with approximately 40%–60% of capsular ruptures occurring during this stage.5,7,8 Most likely, this complication is owing to the combination of ultrasound energy and the sharp needle tip. While outcomes after capsular rupture are typically good,2–4,6–8 there is a definite increase in the incidence of temporarily raised intraocular pressure, persistent uveitis, endophthalmitis, cystoid macular edema, retinal detachment, and retained nuclear material, complications that in some cases necessitate additional surgery.2,3,7,8 Thus, many advances in phacoemulsification technique and technology have centered on reducing the risk of capsular rupture to increase the safety of the procedure. Given the wide array of options available to the cataract surgeon regarding types of vacuum pumps, needle specifications, and ultrasound modalities, little data exist comparing the safety profiles of these different options. Recent studies have evaluated optimization of phacoemulsification variables for maximum efficiency of nuclear emulsification and decreased chatter9–12; however, all of those optimized configurations were found with the machines in peristaltic mode. A recent study by our group showed that Venturi vacuum significantly improved the efficiency of lens fragment removal when compared to peristaltic vacuum.13 Owing to the inherent differences in vacuum generation and fluid dynamics between peristaltic and Venturi pumps, their safety profiles cannot be assumed to be identical. This study assesses the risk of rupture if the phacoemulsification needle were to inadvertently contact the capsule during cataract surgery, and compares optimized phacoemulsification settings in peristaltic and Venturi mode.
METHODS DESIGN:
This study was designed as an experimental laboratory investigation. No human or animal studies were done as part of this investigation.
ELSEVIER INC. ALL
RIGHTS RESERVED.
179
PREVIOUS WORK:
Our methods and experimental design are a continuation and expansion of work by Meyer and associates.14 This previous study used peristaltic modes on the Infiniti (Alcon, Inc, Fort Worth, Texas, USA) and Signature (Abbot Medical Optics [AMO], Inc, Santa Ana, California, USA) phacoemulsification machines with a bottle height of 75 cm, machine-indicated flow rate of 60 mL/min, and 550 mm Hg vacuum setting. Their micropulsed ultrasound trials used a WhiteStar handpiece (AMO, Inc) at 2 longitudinal power settings, 10% and 100%, both with a 6 ms on-duty cycle and 12 ms off-duty cycle. Their transversal ultrasound trials used an Ellips handpiece (AMO, Inc) at 100% power and torsional ultrasound trials used an OZil handpiece (Alcon, Inc) at 100% power. For each of these 4 ultrasound modalities, they tested 4 needles: (1) 19 gauge sharp, (2) 19 gauge Dewey Radius (MST, Redmond, Washington, USA), (3) 20 gauge sharp, and (4) 20 gauge Dewey Radius. For each condition, the tip was tapped against the capsule surrogate 200 times. Additionally, they tested 20 fresh human cadaver lenses with exactly the same phacoemulsification parameters. However, owing to the limited number of cadaver lenses, they were limited to the following conditions: (1) micropulsed ultrasound at 100% power with a 6 ms on-duty cycle and 12 ms off-duty cycle with a 20 gauge Dewey Radius needle, (2) transversal ultrasound at 100% power with a 20 gauge sharp needle, (3) transversal ultrasound at 100% power with a 20 gauge Dewey Radius needle, and (4) torsional ultrasound at 100% power with a 20 gauge Dewey Radius needle. The lenses were tapped gently in different places until capsule rupture occurred, rendering the lenses unusable for further trials.14 They found the rounded edge tip to be very protective of the capsule with all modalities tested.
CAPSULE SURROGATE:
Plastic wrap (Great Value Clear Plastic Wrap; Wal-Mart Stores, Inc, Bentonville, Arkansas, USA) was stretched tightly over one end of a closed 4-inch-diameter polyvinylchloride tube as a surrogate for the capsule in the manner previously described by Meyer and associates.14 The Meyer study used fresh whole human cadaver lenses to further validate this approach. All trials were performed on plastic wrap from the same roll, and pieces of plastic wrap were cut to be large enough to extend beyond the edge of the tube by at least 1 inch in all directions. Tension on the capsule surrogate was established by placing it over the pipe underwater to ensure no air bubbles were trapped underneath, then placing a rubber band around the diameter of the tube to secure it. Once secured in this fashion, to ensure a reasonably uniform tension in all directions, the capsule surrogate was examined for wrinkling, as this would indicate uneven tension. Any wrinkling was resolved by adjusting tension in the appropriate axis of the plastic wrap. This adjustment process was repeated until the entire surface of the capsule surrogate was smoothly uniform. This formed a chamber that was 180
completely filled with water and then submerged in a water bath in order to have a fluid interface on both sides of the capsule surrogate. PHACOEMULSIFICATION
Signature SETTINGS: The (AMO, Inc) phacoemulsification machine was used with the Fusion cassette (AMO, Inc), with all trials using a 550 mm Hg vacuum setting in both peristaltic and Venturi modes, a bottle height of 50 cm, and balanced salt solution. All trials in peristaltic mode used a flow rate of 40 mL/min. While we do not know the exact fluid flow for Venturi vacuum at 550 mm Hg, in a previous study at 500 mm Hg fluid flow was 102.2 6 2.3 mL/min for micropulsed longitudinal and 96.6 6 0.7 mL/min for transversal ultrasound.13 Micropulsed ultrasound trials used a WhiteStar handpiece (AMO, Inc), while transversal ultrasound trials used an Ellips FX handpiece (AMO, Inc). Micropulsed ultrasound was set at 50% power with a 6 ms on-duty cycle and 6 ms off-cycle.15,16 Transversal ultrasound was set at continuous 50% power. Ultrasound and vacuum were both used at only their maximum setting, with all settings on panel control to ensure uniformity. While previous work9 has shown that micropulsed ultrasound is most efficient at 100% power and transversal at 50% power, we decided the fairest test would be to use the same power setting for both in order to minimize power settings as a confounding factor if there were differing breakage rates.
PHACOEMULSIFICATION TIPS:
All trials used phacoemulsification tips with a 30 degree bevel and 0.9 mm (20 gauge) tip diameter. Transversal ultrasound trials were done with a 30 degree bent (Kelman) tip, while micropulsed trials used a straight tip. All tips were made by MicroSurgical Technology (Redmond, Washington, USA). A new tip was used at the beginning of each condition to control for the effect of repeated contact with the capsule surrogate on tip sharpness.
CONTACT WITH CAPSULE SURROGATE:
In order to distinguish possible differences in safety between peristaltic and Venturi pump modes, 2 techniques for contacting the capsule surrogate simulated different clinical situations. The first technique consisted of tapping the edge of the tip against the plastic wrap by approaching the membrane with the needle bevel up at an angle approximately 30 degrees above parallel to the testing membrane, with no tip occlusion, as indicated in the Figure, Left (tip contact). Ultrasound was engaged during the approach to the surrogate capsule, needle contact, and retraction away from the membrane. The second technique involved tapping the needle opening at an appropriate angle against the plastic wrap to achieve tip occlusion. To achieve proper and even occlusion, the handle was held such that the bevel of the needle was facing downward with the needle opening parallel to the capsule surrogate as indicated in the Figure,
AMERICAN JOURNAL OF OPHTHALMOLOGY
JULY 2015
FIGURE. Two approaches were used to contact a surrogate of the human lens capsule with a phacoemulsification needle, in order to assess the risk of capsule rupture using peristaltic and Venturi vacuum pumps with micropulsed and transversal ultrasound. The line represents the capsule surrogate. (Left) Tip contact: The needle is held bevel up, at an angle approach approximately 30 degrees from the capsule surrogate. (Right) Tip occlusion: The needle is held bevel down, with the needle bevel opening parallel to the capsule surrogate.
Right (tip occlusion). Ultrasound was engaged during approach to the surrogate and contact and establishment of occlusion; once tip occlusion was indicated by the increasing pitch of audio tones, which signaled increasing vacuum, the pedal was released fully and the tip was simultaneously retracted away from the plastic wrap. The total time of occlusion from initial contact of the needle to release and retraction away from the membrane was approximately 1 second. In both techniques, contact between the needle and capsule surrogate was monitored under direct visualization, which was aided by placement of a light that would reflect off the membrane, allowing observation of the time and point of contact. Eight conditions were tested: (1) micropulsed ultrasound in peristaltic mode with tip contact; (2) micropulsed ultrasound in peristaltic mode with tip occlusion; (3) transversal ultrasound in peristaltic mode with tip contact; (4) transversal ultrasound in peristaltic mode with tip occlusion; (5) micropulsed ultrasound in Venturi mode with tip contact; (6) micropulsed ultrasound in Venturi mode with tip occlusion; (7) transversal ultrasound in Venturi mode with tip contact; and (8) transversal ultrasound in Venturi mode with tip occlusion. For each occlusion condition, contact was made with the plastic wrap in different places for a total of 200 taps. For each tip contact, contact was made with the plastic wrap in different places for a total of 400 taps. This differential in the number of taps was owing to the considerable difference in breakage rates between tip occlusion and tip contact conditions that we observed in preliminary experiments. ANALYSIS OF CAPSULE SURROGATE:
After each run of 200 or 400 taps, the plastic wrap was examined under a microsurgical microscope (Leica Microsystems Inc, Buffalo Grove, Illinois, USA) using a Sinskey hook (Bausch & Lomb Inc, Rochester, New York, USA) to gently probe
VOL. 160, NO. 1
for breakage. Additionally, in tip occlusion trials we further verified that occlusion had been achieved by observing a full 360 degree outline of the needle in the plastic wrap for each trial. A x2 analysis was used to compare breakage rates between conditions, looking at differences between peristaltic and Venturi modes; between micropulsed and transversal ultrasound; and between tip edge contact and tip occlusion conditions. Results were considered statistically significant at P < .007 after a Bonferroni correction for multiple comparisons.
STATISTICAL ANALYSIS:
RESULTS SEVEN COMPARISONS WERE PERFORMED ON THE DATA SET:
(1) peristaltic vs Venturi pump modes with occlusion data included, (2) peristaltic vs Venturi pump modes without occlusion data, (3) micropulsed vs transversal ultrasound with occlusion data included, (4) micropulsed vs transversal ultrasound without occlusion data, (5) tip contact vs tip occlusion techniques, (6) tip contact using micropulsed ultrasound in peristaltic vs Venturi pump modes, and (7) tip contact using transversal ultrasound in peristaltic vs Venturi pump modes. We analyzed the peristaltic vs Venturi pump data and the micropulsed vs transversal ultrasound data twice, including the occlusion data in one comparison and excluding it in another. This is because with the occlusion techniques, in all 800 trials and regardless of ultrasound modality or vacuum pump type, not a single break occurred. Hence, in comparisons where the tip occlusion data were included, the large number of included trials that did not cause any breaks resulted in a smaller percentage difference
SAFETY OF VENTURI VERSUS PERISTALTIC PHACOEMULSIFICATION PUMPS
181
TABLE. Breaks in a Capsular Surrogate for the Human Lens for Peristaltic and Venturi Phacoemulsification Pumps, Using Micropulsed and Transversal Ultrasound Modalities, and Tip Contact or Tip Occlusion Techniques, With Machine Set at 550 mm Hg, 50 cm Bottle Height (Actual), and When in Peristaltic Mode a 40 mL/min Flow Rate (Machine Indicated) Conditions
Vacuum Pump Type
Ultrasound Modality
Tip Contact or Occlusion
1 2 3 4 5 6 7 8
Peristaltic Peristaltic Peristaltic Peristaltic Venturi Venturi Venturi Venturi
Micropulsed Micropulsed Transversal Transversal Micropulsed Micropulsed Transversal Transversal
Contact Occlusion Contact Occlusion Contact Occlusion Contact Occlusion
No. of Trials Performed
No. (%) of Breaks
400 200 400 200 400 200 400 200
246 (61.5%) 0 284 (71%) 0 208 (52%) 0 274 (68.5%) 0
P < .0001 for all tip contact vs occlusion (rows 1, 3, 5, & 7 vs 2, 4, 6, & 8). P ¼ .013 peristaltic vs Venturi for all comparisons with contact (rows 1 & 3 vs 5 & 7). P < .0001 for all micropulsed vs transversal with contact (rows 1 & 5 vs 3 & 7). P ¼ .0067 for micropulsed contact only peristaltic vs Venturi (row 1 vs 5). P ¼ .44 for transversal contact only peristaltic vs Venturi (row 3 vs 7).
in breakage rates. Thus, in an effort to unmask possible smaller differences between variables, we analyzed the data both ways. We found that there was no significant difference in breakage rates between peristaltic and Venturi pump modes. When tip occlusion data were included in this comparison, breakage rates were 44.2% for peristaltic pump mode and 40.2% for Venturi pump mode (P ¼ .047). When tip occlusion data were excluded from this comparison, breakage rates were 66.3% for peristaltic pump mode and 60.3% for Venturi pump mode (P ¼ .013). Neither of these comparisons exceeded our corrected level of statistical significance. We further found that transversal ultrasound was significantly more likely to cause a breakage than micropulsed ultrasound. When tip occlusion data were included in this comparison, breakage rates were 46.5% for transversal ultrasound and 37.8% for micropulsed ultrasound (P < .0001). When tip occlusion data were excluded from this comparison, breakage rates were 69.8% for transversal ultrasound and 56.8% for micropulsed ultrasound (P < .0001). Both these comparisons exceeded our corrected level of statistical significance. As stated above, not a single break occurred in any of our tip occlusion trials, regardless of ultrasound modality or vacuum pump type. Thus, contacting the capsular surrogate with the tip was significantly more likely to cause a breakage than tip occlusion, with a breakage rate of 63.3% for tip contact and 0% for tip occlusion (P < .0001) (Table). In an attempt to isolate the effect of switching between peristaltic and Venturi vacuum pump modes, an additional 2 comparisons were identified and analyzed. First, breakage rates with tip contact using micropulsed ultrasound were 61.5% for peristaltic pump mode and 52% using Venturi 182
pump mode, which reached statistical significance (P ¼ .0067). Second, breakage rates with tip contact using transversal ultrasound were 71% for peristaltic pump mode and 68.5% for Venturi pump mode, which was not significant (P ¼ .4415).
DISCUSSION CAPSULE RUPTURE MOST COMMONLY OCCURS DURING THE
phacoemulsification stage of cataract surgery. Although surgeons make every attempt to avoid contacting the capsule with the phacoemulsification needle, occasional inadvertent contact is inevitable. Meyer and associates14 observed that 6 variables have an effect on likelihood of capsule breakage: (1) the amount of pressure of the tip against the capsule; (2) the amount of active vacuum at the tip; (3) the speed of flow of the machine; (4) the gauge of the needle; (5) the needle sharpness and degree of angulation; and (6) the energy modulation active at the time of contact. Our study examined the second of these variables, which results from the inherent differences between peristaltic and Venturi pumps in fluidics at the phacoemulsification tip. The results of this study suggest 2 additional variables: (7) the ultrasound modality in use; and (8) the angle at which the tip contacts the capsule. Many surgeons have observed both anecdotally and experimentally that use of Venturi pumps can decrease surgical time, but at the cost of higher postocclusion surge and shorter time to react to events that occur during phacoemulsification and aspiration of cataractous nuclear material.17–20 Contrary to expectations and paradoxically, our results show that the Venturi pump may have a marginally lower risk of capsular breakage than the
AMERICAN JOURNAL OF OPHTHALMOLOGY
JULY 2015
peristaltic pump, although this difference did not quite reach statistical significance (P ¼ .0128). Once we included tip occlusion data in the comparison, the difference between breakage rates for peristaltic and Venturi systems clearly was not significant. One possible explanation is that unoccluded flow vacuum in peristaltic systems is greater than some might assume.14,18,21,22 This would decrease the anticipated difference between the 2 systems in vacuum produced at the phacoemulsification tip, even without occlusion. It is difficult to explain why peristaltic vacuum might be worse. However, we believe our results support the conclusion that at the very least, there is no significant difference between Venturi and peristaltic pumps in risk of capsular breakage. Our analysis revealed 2 other notable findings, both of which achieved statistical significance. The first was the difference in breakage rates between transversal and micropulsed ultrasound modalities, with significantly higher breakage rates in the transversal ultrasound trials. This finding persisted regardless of whether the tip occlusion data were included or excluded. In this case, it may be that the transversal motion produces more of a sideways ‘‘slicing’’ effect on the membrane, while the traditional micropulsed ‘‘jackhammer’’ motion simply impacts and stretches the membrane with less likelihood of rupturing it. Secondly, we found that contact of the tip edge with the membrane was significantly more likely to cause breakage than with full occlusion of the tip by the membrane. Indeed, during a total of 800 trials with both ultrasound modalities and both vacuum pump types in which we produced occlusion, not a single break occurred. Given the relatively high vacuum settings used in this experiment, this result is surprising, particularly because of the theoretically higher unoccluded flow vacuum that the Venturi pump creates. This finding, in conjunction with our initial observation that the Venturi and peristaltic modes likely had fairly similar breakage rates, suggests that the main potential for phacoemulsification-associated capsular rupture may be a consequence of the inherent sharpness of the needle. The Meyer study, which documented that a radiused edge on the phacoemulsification needle, where all sharpness was removed, dramatically reduced the risk of capsular breakage, provides further support of this possibility. Thus, it is also possible that the magnitude of this effect may differ based on the variability in tip finishes from different needle manufacturers or differences in the edge over time, especially small barbs. More investigation, especially of torsional motion, is needed to further explore these findings. Our isolated peristaltic pump vs Venturi pump comparison revealed an interesting contrast between micropulsed and transversal ultrasound. Namely, when tip occlusion data were excluded and only tip contact data were considered, with micropulsed ultrasound there was a statistically significant difference in breakage rates, while with transVOL. 160, NO. 1
versal ultrasound the difference in breakage rates was not significant. One possible explanation for this difference rests with the different movements present at the needle tip in each ultrasound modality. As we conjectured above, the higher breakage rates seen in transversal ultrasound may be attributed to a sideways ‘‘slicing’’ effect not present in the micropulsed ultrasound modality. It may be that in the presence of a larger risk factor for capsule rupture, in this case transversal ultrasound, the difference made in breakage rates by the vacuum pump type becomes less of a factor. Contrast this with micropulsed ultrasound, in which the ‘‘jackhammer’’ motion lowers the risk of capsule rupture but unmasks the more subtle risk of differing vacuum pumps. As with our prior discussion above, it is difficult to explain why the peristaltic pump might be worse; however, we should note that this finding only exceeded our significance cutoff by a very small margin and thus should not be given the same weight as our other findings. Additionally, these comparisons were done excluding tip occlusion data, magnifying differences between other variables. We again emphasize that at the very least our results clearly support the conclusion that there is no significant difference in the risk of capsular breakage between Venturi and peristaltic pumps. From a clinical perspective, several important conclusions can be drawn from our results and put into practice. Most clearly demonstrated is the difference in breakage rates between tip contact and tip occlusion. In surgical practice, this most closely mirrors the ‘‘bevel-up’’ and ‘‘bevel-down’’ approaches. Contrary to common belief, our data support a bevel-down approach to phacoemulsification from the standpoint of reducing the risk of capsule rupture. This is because with a bevel-down approach, in the event of inadvertent contact with the capsule the surgeon would increase the likelihood of a tip occlusion–type event and decrease the likelihood of a tip contact–type event. Furthermore, we demonstrated that using a Venturi vacuum pump in cataract surgery does not increase the risk of capsule rupture. While our data do not address some of the common comparisons of Venturi systems having higher postocclusion surge and shorter reaction time intraoperatively compared to peristaltic systems, they do support the conclusion that Venturi and peristaltic vacuum pumps can be considered equivalent in their risk of capsule rupture. Thus, surgeons should consider other variables, such as efficiency, effective phacoemulsification time, and personal preference, when evaluating vacuum pump types. The principal limitation of this study is that our methods are simulations of situations that can occur during cataract surgery. It would not be possible to conduct such an experiment using human subjects. Additionally, a study with methods similar to ours14 used a relatively small sample of human cadaver capsules in addition to the capsular surrogate; their results from the cadaver capsules were consistent with what they observed using plastic wrap as a
SAFETY OF VENTURI VERSUS PERISTALTIC PHACOEMULSIFICATION PUMPS
183
capsular surrogate. Our results would be prohibitively expensive to investigate and replicate using human cadaver capsules, and we believe that our methods and procedures provide a feasible and effective way to accomplish our objectives. At the very least the risk ratio is likely to be similar, even if the absolute risk is likely to be somewhat different. In conclusion, it is important to note that the risk of capsule rupture is directly related to direct contact with
the capsule by the sharp point of the tip and is not secondary to occlusion even with ultrasound on. Because of this finding, we did not see a significant difference in capsular contact risk between Venturi and peristaltic vacuum even when we used very aggressive parameters. The use of transversal ultrasound appears to increase capsular rupture risk, and our previous work has shown that edge sharpness of the tip is an additional risk factor. Other tip motions need to be assessed in an equally rigorous fashion.
ALL AUTHORS HAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST. Mr Gilbert, Dr Zaugg, Dr Stagg, and Dr Olson do not have any proprietary or financial interest, or any other conflicts of interest, to report. This study was supported in part by an unrestricted grant from Research to Prevent Blindness, Inc, New York, New York, USA, to the Department of Ophthalmology and Visual Sciences, University of Utah, Salt Lake City, Utah, USA. Dr Stagg and Dr Zaugg are recipients of Achievement Rewards for College Scientists Foundation, Inc (ARCS), Utah Chapter, Salt Lake City, Utah, USA, Scholar Awards. MicroSurgical Technology (Redmond, Washington, USA) donated the tips that were used for the study. All authors attest that they meet the current ICMJE requirements to qualify as authors. Susan Schulman, University of Utah School of Medicine, Salt Lake City, Utah, served as a consulting medical writer.
REFERENCES 1. Pandey SK, Milverton EJ, Maloof AJ. A tribute to Charles David Kelman MD: ophthalmologist, inventor and pioneer of phacoemulsification surgery. Clin Experiment Ophthalmol 2004;32(5):529–533. 2. Ang GS, Whyte IF. Effect and outcomes of posterior capsule rupture in a district general hospital setting. J Cataract Refract Surg 2006;32(4):623–627. 3. Chan FM, Mathur R, Ku JJ, et al. Short-term outcomes in eyes with posterior capsule rupture during cataract surgery. J Cataract Refract Surg 2003;29(3):537–541. 4. Ionides A, Minassian D, Tuft S. Visual outcome following posterior capsule rupture during cataract surgery. Br J Ophthalmol 2001;85(2):222–224. 5. Pingree MF, Crandall AS, Olson RJ. Cataract surgery complications in 1 year at an academic institution. J Cataract Refract Surg 1999;25(5):705–708. 6. Wilczy nski M, Wilczy nska O, Synder A, Omulecki W. Incidence and functional outcome of phacoemulsification complicated by posterior capsular rupture. Klin Oczna 2009; 111:26–29. 7. Ti SE, Yang YN, Lang SS, Chee SP. A 5-year audit of cataract surgery outcomes after posterior capsule rupture and risk factors affecting visual acuity. Am J Ophthalmol 2014;157(1): 180–185. 8. Osher RH, Cionni RJ. The torn posterior capsule: its intraoperative behavior, surgical management, and long-term consequences. J Cataract Refract Surg 1990;16(4):490–494. 9. DeMill DL, Zaugg BE, Pettey JH, et al. Objective comparison of 4 nonlongitudinal ultrasound modalities regarding efficiency and chatter. J Cataract Refract Surg 2012;38(6):1065–1071. 10. Oakey ZB, Jensen JD, Zaugg BE, Radmall BR, Pettey JH, Olson RJ. Porcine lens nuclei as a model for comparison of 3 ultrasound modalities regarding efficiency and chatter. J Cataract Refract Surg 2013;39(8):1248–1253. 11. Farukhi AM, Stagg BC, Ronquillo C Jr, et al. Effect of phaco tip diameter on efficiency and chatter. J Cataract Refract Surg 2014;40(5):811–817.
184
12. Gupta I, Zaugg B, Stagg BC, et al. Phacoemulsification efficiency with a radiused phaco tip. J Cataract Refract Surg 2014;40(5):818–821. 13. Cahoon JM, Gupta I, Gardiner G, et al. A comparison between Venturi and peristaltic vacuum in phacoemulsification. J Cataract Refract Surg 2015;41(2):428–432. 14. Meyer JJ, Kuo AF, Olson RJ. The risk of capsular breakage from phacoemulsification needle contact with the lens capsule: a laboratory study. Am J Ophthalmol 2010;149(6): 882–886. 15. Kirk KR, Ronquillo C Jr, Jensen JD, et al. Optimum on-time duty cycle for micropulse technology. J Cataract Refract Surg 2014;40(9):1545–1548. 16. Jensen JD, Kirk KR, Gupta I, et al. Determining optimal ultrasound off time with micropulsed longitudinal phacoemulsification. J Cataract Refract Surg 2015;41(2):433–436. 17. Payne M, Georgescu D, Waite AN, Olson RJ. Phacoemulsification tip vacuum pressure: comparison of 4 devices. J Cataract Refract Surg 2006;32(8):1374–1377. 18. Floyd MS, Valentine JR, Olson RJ. Fluidics and heat generation of Alcon Infiniti and Legacy, Bausch & Lomb Millennium, and Advanced Medical Optics Sovereign phacoemulsification systems. Am J Ophthalmol 2006;142(3): 387–392. 19. Georgescu D, Payne M, Olson RJ. Objective measurement of postocclusion surge during phacoemulsification in human eye-bank eyes. Am J Ophthalmol 2007;143(3):437–440. 20. Patricio MS, Almeida AC, Rodrigues MP, Guedes ME, Ferreira TB. Correlation between cataract grading by Scheimpflug imaging and phaco time in phacoemulsification using peristaltic and Venturi pumps. Eur J Ophthalmol 2013; 23(6):789–792. 21. Georgescu D, Kuo AF, Kinard KI, Olson RJ. A fluidics comparison of Alcon Infiniti, Bausch & Lomb Stellaris, and Advanced Medical Optics Signature phacoemulsification machines. Am J Ophthalmol 2008;145(6):1014–1017. 22. Brinton JP, Adams W, Kumar R, Olson RJ. Comparison of thermal features associated with 2 phacoemulsification machines. J Cataract Refract Surg 2006;32(2):288–293.
AMERICAN JOURNAL OF OPHTHALMOLOGY
JULY 2015
Biosketch Michael Gilbert, MD, graduated from the Creighton University School of Medicine. He is currently an intern at the University of Kansas Medical Center and will be continuing there as a resident in the Department of Ophthalmology in 2016.
VOL. 160, NO. 1
SAFETY OF VENTURI VERSUS PERISTALTIC PHACOEMULSIFICATION PUMPS
184.e1
©2015 Elsevier