Download Corneal Hydration Intra-operatively During Phacoemulsification

249
December 2005
Original Article
Corneal Hydration Intra-operatively During Phacoemulsification
Abhay R. Vasavada, MS, FRCS; Sheena A. Dholakia, DO
Purpose: To evaluate incidence and risk factors for corneal hydration intra-operatively during
phacoemulsification (CHIP).
Methods and Materials: This was a randomised, clinical trial, where 240 eyes of 240 patients with senile
cataracts undergoing phacoemulsification were prospectively randomised to receive one of three possible
types of incision architecture – single, two or three planes. Viscoelastic injection through paracentesis,
before instruments were withdrawn from the main incision, was randomly done. Incision length was
grouped into 1.5, 2 and 2.5 mm. CHIP was graded as nil, mild, moderate or severe: nil - no visible CHIP,
mild-CHIP involving pillars of the incision, moderate - CHIP involving pillars and lateral extension, severe
- extension of CHIP in front into the clear cornea. Moderate and severe categories were combined as
‘significant’ CHIP. Statistical analysis was done by logistic regression.
Results: The incidence of significant CHIP was 17.9 % (43 eyes out of 240). Five out of 80 eyes (6.25%)
single plane, 8 out of 80 (10%) two plane and 30 out of 80 (37.5%) three- plane incisions developed
significant CHIP (p = 0.005). Fifteen out of 120 (12.5%) with and 28 out of 120 (23.3%) without injection of
viscoelastic developed significant CHIP (p = 0.044). Eighteen out of 43 (42.8%) and 25 out of 43(58.1%)
with incision lengths 2 and 2.5 mm developed significant CHIP respectively (p<0.001).
Conclusions: The incidence of significant CHIP was 17.9 %. Three-plane incisions, not injecting viscoelastic
prior to retraction of instruments and incision length 2 mm or more are risk factors for significant CHIP.
Key words: incision, phacoemulsification, viscoelastic substance
Indian J Ophthalmol 2005;53:248-253
Clear corneal incisions have become the preferred
technique in phacoemulsification.1,2 Clear corneal incisions
have several disadvantages, including external deformation,
delayed wound healing and increased risk of
endophthalmitis.1,3,4
At our training centre, we have occasionally observed the
phenomenon of intra-operative corneal hydration during
phacoemulsification. We have termed this condition as CHIP
–
Corneal
hydration
intra-operatively
during
phacoemulsification. This condition does not refer to the
corneal hydration performed deliberately at the end of surgery
for better sealing of the incision.1,5
The purpose of the study was to evaluate patients
undergoing phacoemulsification to determine incidence,
severity and risk factors which may cause CHIP.
Material and Methods
This is a prospective, randomised, observational case series
of 240 consecutive eyes of 240 patients undergoing
Iladevi Cataract & IOL Research Centre, Raghudeep Eye Clinic,
Gurukul Road, Memnagar, Ahmedabad, India
Correspondence to Dr. Abhay Vasavada, Iladevi Cataract & IOL
Research Centre, Raghudeep Eye Clinic, Gurukul Road, Memnagar
Ahmedabad India. E-mail: <[email protected]>
Manuscript received: 13.4.05; Revision accepted: 25.7.05
phacoemulsification at Iladevi Cataract and IOL Research
Centre, Ahmedabad, India between September 2004 and
January 2005. They were recruited in a continuous cohort, after
informed consent of patients.
The sample size was calculated using the formula
n=
2pq [Zα - Zβ]2
(p - q - δ∗)2
≅ 1
≅ 200
0.0049
p = q = 50,
δ* = 0.07,
α = 0.05,
β = 0.20.
The element of chance was set to 5% and power was set to
80% for calculation of sample size. The allowable error was
7%, as we wanted an error margin of less than 10% and more
than 5%. An additional 20% observations were taken to tackle
data problems.
All types of senile cataract ranging in an otherwise normal
eye were included in the study. Patients with shallow anterior
chamber (<2.5 mm) and patients with diabetes or hypertension
were also included.
Total, sub-luxated, posterior polar cataracts and patients
249 CMYK
INDIAN JOURNAL OF OPHTHALMOLOGY
The recruited patients were seen preoperatively by a single
observer (SAD), through the slit lamp biomicroscope (SL-120
Carl Zeiss, Germany) after mydriasis, using Tropicamide
(Tropicacyl 1%, Sunways (I) P. Ltd., India) eye drops. The
density of cataract was graded according to the classifcation
of Emery and Little6 using system of grades 1–5. The oblique
slit beam width and height were fixed at 0.3 and 0.9 mm,
respectively, looking at 45 0 and fixing illumination at
intermediate level. Fixation light was set so that the slit-lamp
beam was on the temporal side of right eye and nasal side of
left eye in such a manner that it bisected the lens from 12 to 6
o’clock and focused near the centre of the lens. For statistical
analysis, eyes with only cortical and/or posterior sub-capsular
cataract were considered as grade 0. Preoperatively,
noncontact specular microscopy using the Konan SP 8000
noncon ROBO specular microscope (Konan Medical Inc.,
Japan) was performed.
The surgical procedures were performed by a single
experienced surgeon (ARV), by a standardised technique. The
two factors randomised were architecture of the incision and
viscoelastic injection through paracentesis before instruments
were withdrawn from the main incision. The envelope system
was used for randomisation of eyes, described in the flowchart
[Figure 1]. In this system, prior to commencing the study
sequenced and sealed envelopes were prepared containing
one of the optional surgical procedures. A nurse in the
operating room would open an envelope just before the
surgery and announce the procedure to the surgeon.
Three types of temporal clear corneal incisions were
constructed depending upon the randomisation. Single-plane
incision was constructed by direct entry with a 3.0 mm angled
slit knife (SatinSlit Alcon surgical). Two-plane incisions was
constructed with 15 No. D knife on a Bard Parker handle with
a groove of approximately half thickness, followed by the entry
with a 3.0 mm angled slit knife (SatinSlit Alcon surgical).
Three-plane incisions had a groove which was made as
A
Vol. 53 No. 4
described above followed by blunt dissection with an angled
bevel up crescent knife (Satin Crescent Alcon surgical) after
which entry was performed with a 3.00 mm angled slit knife
(SatinSlit Alcon surgical) [Figure 2A–C].
The anterior chamber was filled with 2% Hydroxy propyl
methyl cellulose (Viscomet, Milmet Inc., India). Subsequently,
continuous curvilinear capsulorhexis (CCC)7 was performed
with 26 G bent cystitome. The paracentesis was made with a
15° Ophthalmic knife (Alcon surgical). Hydrodissection8 was
performed with a 27 G cannula mounted on a 2 cc syringe.
The nucleus was rotated with an iris spatula through the
paracentesis.
Phacoemulsification was performed with an Alcon® Legacy
20 000 phacoemulsifier. Central sculpting was done with 25°
bent kelman microtip with a 450 bevel. Parameters were
adjusted appropriate to the grade of cataract. For firm
cataracts, above grade-1, we used our previously published
technique ‘step by step, chop in situ and lateral separation’9
and subsequently removed it by the chop, chop and stuff
technique.10 For soft cataracts, we performed nucleotomy and
then aspirated the lens substance using the Alcon Legacy
20000TM phacoemulsifier.
The cortex was aspirated with a single port Irrigation/
Aspiration (0.3 mm diameter) cannula. The difficulty in
Assessed for eligibility (n=400 eyes)
Enrolment
with a single functioning eye were excluded. Patients with
co-existing corneal pathology (such as old healed viral
keratitis, eyes with vascularised scar, corneal dystrophy and
degeneration), other ocular diseases (such as uveitis, glaucoma
and pseudoexfoliation), history of intraocular surgery (such
as anti-glaucoma surgery, retinal detachment surgery and
refractive corneal surgery) and small pupils were also
excluded.
Excluded (n = 160 eyes)
Not meeting inclusion criteria (n= 160 eyes)
Randomization (n =240 eyes)
Allocation and Analysis
250
with injection of viscoelastic
before retracting instruments
(n=120 eyes)
without injection of viscoelastic
before retracting instruments
(n=120 eyes)
Single
plane
(n=40)
Single
plane
(n=40)
Two
plane
(n=40)
Three
plane
(n=40)
(None excluded from analysis)
Two
plane
(n=40)
Three
plane
(n=40)
(None excluded from analysis)
Figure 1: Flow chart
B
C
Figure 2: A Depicts graphic representation of single-plane incision in profile. B Depicts graphic representation of two-plane incisions in profile. C
Depicts graphic representation of three-plane incisions in profile
250 CMYK
December 2005
Vasavada, et al. Corneal hydration intra-operatively during phacoemulsification
Figure 3: Incisional distortion. CHIP was observed during the irrigationaspiration stage due to distortion of the incision
A
251
Figure 4: Viscoelastic injection. Injection of viscoelastic substance
through paracentesis prior to retraction of instruments from the main
incision showing no CHIP
B
C
Figure 5: A Depicts mild stage of CHIP. B Depicts moderate stage of CHIP. C Depicts severe stage of CHIP
A
B
C
Figure 6: A CHIP during the stage of sculpting. Note that CHIP is mild. B CHIP during the stage of fragment removal. Note that CHIP has
progressed. C CHIP during the stage of irrigation-aspiration. Note the further progression of CHIP
removing sub-incisional cortex and the consequent distortion
of the incision was subjectively graded by the surgeon and
categorised into distortion of incision [Figure 3] and no
distortion of incision.
randomly assigned [Figure 4]. Subsequently an Alcon Acrysof®
SA60AT was implanted in the capsular bag. Residual Viscomet®
was removed from the anterior chamber and from behind the
IOL. Vancomycin (0.1 mg/1 ml) was injected intra-camerally.
Viscoelastic injection through paracentesis before
instruments were withdrawn from the main incision was
The length of the incision (between external and internal
lips) was grouped into three as 1.5, 2 and 2.5 mm. The incisions
251 CMYK
252
INDIAN JOURNAL OF OPHTHALMOLOGY
were constructed by presetting the mark with a calipers. Total
surgical clock time in minutes was noted.
Primary observations related to CHIP were recorded as nil
– no visible CHIP; mild – CHIP involving only pillars of the
incision; moderate – CHIP involving pillars and/or lateral
extension; severe – extension of CHIP into the clear cornea
[Figure 5A–C]. These observations were made during the
sculpting, division, fragment-removal and irrigationaspiration stages [Figure 6A–C] of the surgery by a single
observer (SAD). Secondary observations of anterior chamber
depth, surgical clock time and endothelial count were recorded
by a single observer (SAD).
Vol. 53 No. 4
Table 2: Logistic regression analysis of CHIP
Variables in the Equation
Log Significance
odds (P Value)
Exp (B) =
odds
Radio
95.0% C.I. for
Odds Ratio
Lower Upper
Step 1Plane
Injection of
1.090
0.005
2.975
1.392
6.358
viscoelastic
1.130
0.044
3.095
1.032
9.288
cortex
-2.012
<0.001
0.134
0.045
0.394
Surgical time
1.178
<0.001
3.248
1.835
5.749
To determine the regression of CHIP, five patients each
with mild, moderate and severe CHIP were followed-up every
8 h till CHIP disappeared.
Length Incision
2.395
<0.001
10.969
3.805 31.618
Logistic regression models using SPSS software were
applied to estimate the odds ratio of occurrence of CHIP. Test
of difference of mean was applied to determine association of
endothelial cell count and CHIP and surgical time and CHIP.
For the purpose of statistical analysis and clinical
interpretation, CHIP was divided into nil and mild CHIP as
one group and moderate and severe CHIP as the second group.
Moderate and severe CHIP was considered to be ‘significant’
CHIP.
Since endothelial count is a running number; it was divided
into less than or equal to median and more than median. Grade
of cataract was divided into less than grade 3 and equal to or
more than grade 3. Length of the incision was classified as
1.5, 2 or 2.5 mm. Total surgical clock time was classified as
<20, 20–30 and >30 min.
Results
None of the eyes subjected to randomisation were excluded
due to intra-operative complications.
The mean age of 128 men was 61.7 years (range 42–71,
median 64 years) and of the 112 women was 60.5 years (range
44–70 years, median 62 years).
The incidence of significant CHIP was 17.9 % (43 eyes out
of 240).
Mild, moderate and severe CHIP occurred in 110 out of
240 (45.8%), 37 out of 240(15.4%) and 6 out of 240 (2.5%), eyes
respectively. Mild and moderate CHIP disappeared within 8
and 32 h, respectively, Severe CHIP disappeared by 48 hours
in 3 eyes and 56 hours in 2 eyes.
Flow chart describes randomisation of surgical
intervention. The stage of surgery at which CHIP developed
is shown in [Table 1]. Logistic regression analysis is described
in [Table 2]. Test of difference of mean to describe correlation
of endothelial cell count and CHIP and surgical time and CHIP
is described in [Table 3].
Table 1: Stage of procedure at which significant CHIP
developed
Sculpting
Chopping
Fragment
removal
Irrigationaspiration
Total
6 (14.0%)
2 (4.6%)
29 (67.4%)
6 (14.0%)
43 (100%)
252 CMYK
Sub-incisional
Endothelial count -0.319
0.555
0.727
0.251
2.1
Grade of cataract 0.861
0.122
2.366
0.794
7.051
Constant
0.00
0.00
-9.972
In the above model, the following factors were statistically associated with
CHIP:
1. Three-plane incisions has 2.9 times the risk of development of CHIP
compared to two-plane incisions, and two-plane incisions has 2.9 times the
risk of development of CHIP as compared to single-plane incision.
2. Not injecting viscoelastic substance prior to retraction of instruments
increases the risk of development of CHIP three times.
3. No sub-incisional cortex difficulty decreases the odds of CHIP by 0.13
times, i.e. 87% lower chance than those having difficulty.
4. Increase in surgical time increases the risk of CHIP by 3.2 times.
5. Incisional length 2.5 mm increases the risk of CHIP by 10.9 times compared
to incisional length of 2 mm, and 2 mm increases the risk of CHIP by 10.9
times as compared to 1.5 mm length.
There is no significant co-relation of endothelial count and grade of cataract.
Table 3: Co-relation of surgical time and endothelial count
and CHIP
Nil + Mild
Moderate +
Severe
P Value of test
of difference of
mean
Mean surgical time
24.6 (5.2)
32.2 (7.9)
0.001
Endothelial count
2059 (257)
2015 (256)
0.31
Above table concludes that the surgical time associated with CHIP is higher
and statistically significant. Although, the endothelial count is lower for CHIP, it
is not statistically significant.
Discussion
The entity of corneal hydration occurring intra-operatively has
been recognised by surgeons, but as far as we are aware, the
incidence and risk factors have not been reported in the
literature. We had observed that CHIP occurs to a variable
degree during phacoemulsification. Therefore, on the basis of
the morphological pattern, we graded it as mild, moderate or
severe. Although, incidence of mild CHIP was high, it does
not hamper the surgeons’ visibility as it is restricted to the
pillars of the incision. On the other hand, moderate and severe
CHIP can obscure visibility and hamper the
phacoemulsification procedure.
Hindrance in the surgeons’ visibility can lead to adverse
consequences such as posterior capsule rupture and vitreous
December 2005
Vasavada, et al. Corneal hydration intra-operatively during phacoemulsification
loss. Because of its clinical relevance, we have considered
moderate and severe CHIP to be ‘significant’. Serious
complications did not occur in the present study partly due
to the extensive experience of the surgeon in performing the
phacoemulsification procedure.
This phenomenon is transient and even in the most severe
cases regression occurred by 48-56 hours. Occasionally,
patients noticed CHIP when it was significant and these few
patients needed reassurance regarding this transient
phenomenon.
Of the incision architecture related factors, we believe that
blunt dissection with a crescent blade produces pockets in
which fluid collects. Hence the incidence of CHIP was highest
in three-plane incisions. Lesser dissection on the sides by the
crescent blade might reduce CHIP.
Viscoelastic placement before instruments were withdrawn
from the main incision has several advantages. Injecting
viscoelastic through the paracentesis has been recommended
by Osher and Cionni to prevent forward bulge of anterior
vitreous and posterior capsule, thus preventing posterior
capsule rupture and/or anterior vitreous prolapse.5
Also, viscoelastic placement eliminates the need to irrigate
while going in and out of the eye. Another indication of this
technique is to prevent internal entry distortion. We believe
that internal entry distortion may be a factor, which may lead
to CHIP. When an instrument is withdrawn from the main
incision, there is a momentary fish mouthing of the internal
entry. Injecting viscoelastic prevents this phenomenon by
maintaining the contours of the globe and creating a
pressurised environment around the internal entry. The
internal entry distortion classically occurs during subincisional manoeuvres such as cortex removal. Bimanual
irrigation aspiration technique11 is preferable when cortex
removal is difficult. In the present study, bimanual irrigation/
aspiration was not performed as the surgical technique was
standardised.
It is interesting to note that ‘significant’ CHIP developed
mainly during the stages of fragment removal and irrigationaspiration. Once the surgeon recognises CHIP in the early
stages, injecting viscoelastic substance through the
paracentesis may prevent further progression.
According to I. Howard Fine (personal communication),
an increase in the surgical time promotes CHIP. The results
of the present study support this view. We presume that
increase in surgical time can be due to a combination of several
factors of which could be grade of cataract, difficulty in
removing sub-incisional cortex and number of instruments
introduced and removed.
Incision length was varied as we believed that longer
incision may lead to CHIP. There is greater distortion of the
internal entry in longer incisions and more corneal stroma
comes in contact with the irrigating fluid.
The corneal endothelial pump mechanism is responsible
for the deturgesence of fluid from the corneal stroma. 12
According to I. Howard Fine (written communication), corneal
endothelial diseases can lead to intra-operative stromal
253
hydration. However, since the process of CHIP is so transient,
the endothelial pump mechanism may not play a significant
role. It is possible that regression may occur later in eyes with
lower endothelial count or with greater endothelial cell loss.
However, we did not examine patients for this phenomenon.
Difficulty in removing sub-incisional cortex was subjectively
evaluated by the surgeon. The lack of objective assessment
can be a source of observer bias.
Several other factors such as energy used during
phacoemuslification and amount of irrigation fluid may
contribute to CHIP. Similarly, although hydorxy propyl
methyl cellulose was the viscoelastic used in the present study,
other viscoelastics may influence the occurrence of CHIP
differently and may be a subject for further research.
This randomised, controlled trial thus concludes that threeplane incisions, not injecting viscoelastic through paracentesis
before instruments were withdrawn from the main incision,
sub-incisional cortex removal difficulty, tunnel length 2 mm
or longer and prolonged surgical time are risk factors of CHIP.
Several of the factors that contribute to CHIP including plane
of incision and incision length are modifiable. Identifying this
phenomenon early in the genesis can help the surgeon take
such appropriate measures as viscoelastic injection through
paracentesis before instruments were withdrawn from the
main incision, reducing the surgical clock time and other
manipulations that could decrease the incidence of CHIP.
References
1.
Fine IH, Hoffman RS. Clear Corneal Cataract Surgery. Ophthal
Surg Lasers 1998;29:822-31.
2.
Leaming DV. Practice styles and preferences of ASCRS
members— 2003 survey. J Cataract Refract Surg 2004;30:892-900.
3.
Barba KR, Samy A, Lai C, Perlman JI, Bouchard CS. Effect of
topical anti-inflammatory drugs on corneal and limbal wound
healing. J Cataract Refract Surg 2000;26:893-7.
4.
Taban M, Behrens A, Newcomb RL, Nobe MY, Saedi G, Sweet
PM, et al. Acute endophthalmitis following cataract surgery: a
systematic review of the literature. Arch Ophthalmol
2005;123:613-20.
5.
Osher RH, Cionni R. Ch. 39, In: Cataract Surgery, Technique,
Complications, Management, 2nd edn. Saunders: Steinert RF;
2004. p. 469–86.
6.
Emery JM, Little JH. Phacoemulsification and aspiration of
cataract, Surgical Technique,Complications and Results. St.
Louis: C.V. Mosby; 1979. p. 45-9.
7.
Gimbel HV, Neuhann T. Development, advantages and
methods of the continuous circular capsulorhexis technique. J
Cataract Refract Surg 1990;16:31-7.
8.
Fine IH. Cortical cleaving hydrodissection. J Cataract Refract
Surg 1992;18:508-12.
9.
Vasavada AR, Singh R. Step-by-step, chop in situ and lateral
separation of very dense cataracts. J Cataract Refract Surg
1998;24:156-9.
10. Vasavada AR, Desai JP. Stop, chop, chop and stuff. J Cataract
Refract Surg 1996;22:526-9.
11. Brauweiler P. Bimanual irrigation / aspiration. J Cataract Refract
Surg 1996;22:1013-6.
12. Smolin G, Thoft RA. Physiology, In: The Cornea. Part I - Basic
Science, Scientific Foundations and Clinical Practice, 3rd edn,
Little, Brown and Company: 1994. p. 25–46.
253 CMYK