Cataract after Vitre.

ctomy in
Young Patients

Barbara A. Blodi, MD, Scott A. Paluska, MD

Objective: The purpose of the study was to evaluate the occurrence of cataract in
young patients after pars plana vitrectomy.
Design: The authors reviewed the medical records for previtrectomy and postvitrec-
tomy lens changes in patients younger than 30 years of age at the Kellogg Eye Center,
University of Michigan.
Participants: Forty-nine patients (50 eyes) younger than 30 years of age (mean
age, 23.5 years; range, 5 months-30 years) underwent phakic vitrectomy over a 12-
year period.
Main Outcome Measures: Cataracts were categorized as posterior subcapsular,
nuclear sclerotic, or cortical. Cataracts also were graded as mild (1 +), moderate (2+),
or severe (3+).
Results: In this series of young patients, vitrectomy was performed for a wide range
of ocular conditions, including trauma and complicated retinal detachment. Postvitrectomy
cataract developed in 29 patients (60%). Eighteen patients (36%) had visually significant
cataract on long-term follow-up (mean follow-up, 29.7 months; range, 6 months-13
years). The most common cataract was posterior subcapsular (57%), followed by nuclear
sclerosis (23%), a combination of both (17%), and cortical cataract (3%). Patients with
gas-filled eyes had a significantly higher rate of cataract formation than patients with
fluid-filled eyes (P < 0.05).
Conclusions: Postvitrectomy cataracts were more common in the authors' series
compared with those of previous reports on young patients. Cataracts were most often
posterior subcapsular and were significantly associated with the use of intraocular gas.
The occurrence of postvitrectomy cataract appears to be higher in patients undergoing
vitrectomy for complex ocular conditions, regardless of age.
Ophthalmology 1997; 104: 1092-1 095

Cataract is a well-known complication of pars plana vi- to 34%5 for posterior subcapsular cataract. Although the
trectomy in phakic patients. 1 - 7 Vitreous surgery is associ- exact cause of cataract formation is unclear, postvitrec-
ated with both the progression of pre-existing cataract tomy cataract has been associated with the following fac-
and the development of new postoperative cataract. In tors: advanced patient age,2.3.6.8 diabetes/.4 length of sur-
the adult population, the reported incidence of new or gery,l prolonged exposure to an intravitreallight source,6,7
progressive lens opacities after vitrectomy ranges from the type of infusion fluid or infusion cannula,5,7,9 and the
12.5%7 to 80%2 for nuclear sclerotic cataracts and 4%3 use of intraocular air, gas, or silicone oiL 6
Information concerning postvitrectomy cataract in
younger patients is sparse, In infants younger than age 3,
Originally received: October 30, 1995. MaguirelO reported a 20% incidence of cataract after lens-
sparing vitrectomy in ten patients, Cherfan et al2 reported
Revision accepted: March 5, 1997.
From Kellogg Eye Center, University of Michigan, Ann Arbor, Michi-
gan.
a 9% incidence of nuclear sclerosis in a subset of 11
patients younger than age 50, Melberg and Thomas ll
Presented at the American Academy of Ophthalmology Annual Meeting,
Atlanta, Georgia, November 1995. found postvitrectomy nuclear sclerosis in 7% of patients
Reprint requests to Barbara A. Blodi, MD, F4/334 Clinical Sciences between the ages of 16 and 50. In the latter two series,
Center, Department of Ophthalmology, University of Wisconsin, 600 the low rate of postvitrectomy nuclear sclerosis occurred
Highland Ave, Madison, WI 53792. in younger patients undergoing macular surgery for epire-

1092
 Blodi and Paluska Cataract after Vitrectomy

Table 1. Age Distribution of Patients Table 2. Distribution of Diabetic Patients

Age (yrs) of Patients Number (%) of Patients Total Number Diabetic Nondiabetic
of Patients [number (%)] [number (%)]
0-9 2 (4)
10-18 6 (12) 49 22 (45) 27 (55)
19-30 41 (84)

tients with diabetes made up 45% of the population (Table

2), and all but one patient with diabetes was older than
tinal membranes or removal of subretinal neovascular 18 years of age. Follow-up ranged from 6 to 156 months
membranes. We hypothesized that the rate of cataract with a mean of 29.7 months.
formation in the pediatric and young adult population The preoperative diagnoses included diabetic traction
may be higher after phakic pars plana vitrectomy for a detachment (13), diabetic vitreous hemorrhage (9), intra-
wider range of ocular conditions. ocular foreign body (7), rhegmatogenous retinal detach-
ment (5), pars planitis (3), cytomegalovirus-related retinal
detachment (3), macular pucker (3), scleral laceration (3),
Methods endophthalmitis (2), Terson syndrome (1), and sickle cell
retinopathy with vitreous hemorrhage (1) (Table 3).
We reviewed all charts and operative reports of patients
30 years of age or younger who underwent phakic pars
plana vitrectomy at the Kellogg Eye Center, University Preoperative Features
of Michigan, between 1982 and 1994. Six vitreoretinal Forty-three patients (86%) had clear lenses before sur-
surgeons performed the vitrectomies over the 12-year pe- gery. Seven patients (14%) had mild (1 +) preoperative
riod. Inclusion criteria consisted of at least 6 months of cataract. Four patients with pre-existing cataract had dia-
follow-up and no prior intraocular surgery. Patients were betes mellitus, two patients had chronic pars planitis, and
excluded if a traumatic cataract or lens injury was present one patient had endophthalmitis. No cataracts were pres-
at the time of vitrectomy. Fifty eyes of 49 patients met ent in the fellow eye of any patient. Before surgery, the
these criteria. two patients with pars planitis had received sub-Tenon's
A detailed chart review provided information regard- steroid injection within 1 month of surgery, whereas the
ing preoperative, intraoperative, and postoperative data three patients with scleral lacerations had a 2-week course
for each patient. Preoperative data included age, gender, of topical steroid drops four to eight times per day. None
medical conditions, affected eye, ocular diagnoses, visual of the patients were receiving oral corticosteroids, immu-
acuity, type and severity of pre-existing lens changes, and nosuppressive agents, or other systemic cataractogenic
status of the fellow lens. Intraoperative factors included medications.
length of surgery, type of infusion fluid, intraoperative
complications, and use of gas, air, or silicone oil. Postop-
Operative Features
erative information consisted of visual acuity, type and
severity of lens changes, need for additional surgery, and In all 50 eyes, a standard 3-port pars plana vitrectomy
primary cause for decreased visual acuity. was performed by a member of the retina service of the
Cataract classification and grading, both before surgery
and after surgery, were performed by the retinal surgeon.
Cataracts were classified as nuclear sclerotic, posterior Table 3. Preoperative Diagnoses
subcapsular, or cortical. The lens opacities were graded
Number (%)
as mild (1 +), moderate (2+), or severe (3+). Patients Diagnosis of Patients
with pre-existing cataract were considered to have pro-
gression if their cataract increased by at least one grade Diabetic traction detachment 13 (26)
or an additional type of cataract developed. Patient infor- Diabetic vitreous hemorrhage 9 (18)
mation was recorded and tabulated on a computerized Intraocular foreign body 7 (14)
Rhegma£Ogenous retinal detachment 5 (10)
database.
Pars planitis 3 (6)
CMV retinal detachment 3 (6)
Macular pucker 3 (6)
Results Scleral laceration 3 (6)
Endophthalmitis 2 (4)
Patient Background T erson syndrome I (2)
Sickle cell retinopathy with vitreous
There were 29 male and 20 female patients with a total hemorrhage 1 (2)
of 50 operated eyes. The mean age of the patients was 50 (100)
Total
23.5 years with a range from 5 months to 30 years. The
pediatric popUlation, defined as 18 years of age or CMV = cytomegalovirus.
younger, consisted of eight patients (16%) (Table 1). Pa-

1093
 Ophthalmology Volume 104, Number 7, July 1997

Kellogg Eye Center. Vitrectomy alone was performed in 0.03, Z test for binomial proportions) developed in only
39 (78 %) of the 50 eyes. In the remaining 11 eyes (22%), 24 (54%) of 44 patients with fluid-filled eyes.
additional procedures were performed. Six patients had a Analysis of other perioperative factors showed that
scleral buckle combined with vitrectomy, four patients postvitrectomy cataract developed in 17 patients with dia-
underwent vitrectomy with scleral laceration repair, and betes (77%) compared with that of 12 patients without
one patient had scleral laceration repair, scleral buckle, diabetes (45%). In addition, the mean surgical time for
and vitrectomy. patients with postvitrectomy cataracts was longer (173
For all vitrectomies, the surgeon placed a 4-mm infu- minutes) than for patients without postoperative cataract
sion cannula between 2.5 mm and 3.5 mm posterior to (146 minutes). Neither the presence of diabetes nor the
the limbus in the inferotemporal quadrant of the operative increased surgical time was statistically significant in the
eye. The intraoperative infusion fluid uniformly consisted development of cataracts (Z test for binomial propor-
of Balanced Salt Solution Plus (BSS Plus; Alcon, Ft. tions). The light source, type of infusion fluid, and posi-
Worth, TX) with 0.25 mll :1000 epinephrine solution per tion of the infusion cannula were constant in all cases.
liter. For diabetic vitrectomies, 300 mg glucose/dl was
added to the infusion fluid. Exact operating time was
recorded for 34 patients (68%); the intraocular surgical Discussion
time ranged from one-half hour to 6 hours with a mean
of 161 minutes. No apparent iatrogenic lens damage oc- In our series of patients younger than 30 years of age,
curred during surgery. cataract after phakic pars plana vitrectomy developed in
At the end of surgery, six patients underwent complete 30 patients (60%). Visually significant lens opacities
fluid-gas exchange; five of six eyes received a 20% con- (moderate to severe) were present in approximately one
centration of sulfur hexafluoride, whereas one eye had air third of all eyes operated on. The incidence of postvitrec-
only. No silicone oil was used as a tamponade. After tomy cataract in young patients in our series differs from
surgery, all patients began a regimen of atropine and a the data of previous authors who conclude that younger
topical steroid and antibiotic combination drop four times patients are less likely to have cataract develop.2,3,6,8 In
a d ay. The drops were tapered between 8 and 16 weeks our series, the rate of visually significant postvitrectomy
after surgery. Long-term topical steroid therapy was not cataract in younger patients is comparable to that of adult
continued with these patients. Postvitrectomy procedures patients.
included lens extraction in six patients. Subsequent vitrec- Interestingly, the proportion of posterior subcapsular
tomy or reinjection of intravitreal gas was not required cataract (57%) developing in younger patients is higher
in any patient. than in adult patients. 3,5,6.12 Postvitrectomy cataracts in
adults are more likely to be nuclear sclerotic. 3.6.8 This is
consistent with the finding in the general population that
Postoperative Features and Cataract Development nuclear sclerosis is more common in adults and that poste-
rior subcapsular cataract is more common in young pa-
Cataract developed or progressed in 30 patients (60%) tients and in patients with diabetes. 13
undergoing phakic pars plana vitrectomy. Of those pa- We found that postvitrectomy cataracts in young adults
tients in whom cataract developed, posterior subcapsular were significantly associated with the use of air or gas as
cataract developed in 17 patients (57%), nuclear sclerotic an intraocular tamponade. In older patients undergoing
cataract developed in 7 patients (23%), five patients vitrectomy for diabetes 6 and rhegmatogenous retinal de-
(17 %) developed both posterior subcapsular and nuclear tachments,12 the use of an intravitreal gas bubble, espe-
sclerotic lens opacities, and cortical cataracts only devel- cially a large gas bubble, was associated with postvitrec-
oped in 1 patient (3%). Cataract developed in two pediat- tomy cataract formation. deBustros et aV reporting on
ric patients (25 %), and in both patients, the lens opacities membrane peeling surgery, did not find a higher rate of
were mild. Cataracts did not develop in the fellow eye of cataract in the small number of cases in which air or
any patient. sulfur hexafluoride gas was used.
In our series, the majority of postvitrectomy cataracts In our series, young patients with diabetes had a higher
appeared early. Of the 30 cataracts, 22 (73%) developed incidence of postvitrectomy cataract than in patients with-
within 6 months, 7 cataracts (23%) developed between 6 out diabetes. In adults, diabetes also accounts for a higher
and 12 months, and 1 cataract (4%) developed after 1 rate of postvitrectomy cataract formation. 1,4 Grewing and
year. Mester4 reported twice as many posterior subcapsular cat-
The severity of the cataract (regardless of type) was aracts after vitrectomy for proliferative diabetic retinopa-
mild in 12 patients (40%), moderate in 10 patients (33%), thy than for macular pucker (33.6% vs. 17.1%). In the
and severe in 8 patients (27%). Visually significant cata- series of Grewing and Mester, patients with diabetes also
racts, defined as moderate to severe, occurred in 18 (60%) had a higher rate of postvitrectomy nuclear sclerosis com-
of the 30 patients with lens opacities. pared to patients with macular pucker (21.8% vs. 14.6%).
In our series, the intraoperative use of air or gas tam- In our series, patients in whom cataract developed after
ponade was correlated strongly with the development of vitrectomy tended to have longer surgeries. This finding ,
postvitrectomy cataract. Cataract developed in all 6 pa- although not statistically significant in our population, is
tients with a gas tamponade, whereas lens opacities (P < similar to data on older patients undergoing vitreous sur-

1094
 Blodi and Paluska . Cataract after Vitrectomy

gery for membrane pucker7 and proliferative diabetic reti- References

nopathy.I,6 Other intraoperative variables, such as the type
of infusion fluid, position of the infusion cannula, and 1. Blankenship G, Cortez R, Machemer R. The lens and pars
type of light source, remained constant in our series and plana vitrectomy for diabetic retinopathy complications.
could not be analyzed. Arch OphthalmoI1979;97:1263-7.
2. Cherfan GM, Michels RG, deBustros S, et al. Nuclear scle-
The development of postvitrectomy cataract appears rotic cataract after vitrectomy for idiopathic epiretinal mem-
to be multifactorial and may relate more to the complexity branes causing macular pucker. Am J Ophthalmol
of the ocular condition than to the age of the patient. 1991; 111:434-8.
For example, in patients with diabetic retinopathy, pars 3. deBustros S, Thompson JT, Michels RG, et al. Nuclear
planitis, trauma, proliferative vitreoretinopathy, or en- sclerosis after vitrectomy for idiopathic epiretinal mem-
dophthalmitis, the development of postvitrectomy cata- branes. Am J Ophthalmol 1988; 105:160-4.
ract may be affected by the underlying disease itself and 4. Grewing R, Mester U. Linsentriibungen nach Pars-plana-
Vitrektomie bei diabetischer Vitreoretinopathie und Macu-
the presence of inflammation. Patients with these condi- lar pucker. Fortschr Ophthalmol 1990;87:440-2.
tions often have longer vitrectomies and a longer postop- 5. Mester U, Grewing R. Linsentriibungen nach pars plana
erative course, requiring the use of high-dose topical cor- vitrektomie. Fortschr Ophthalmol 1989;86:429-32.
ticosteroids.I,4,6 In contrast, patients with epiretinal mem- 6. Novak MA, Rice TA, Michels RG, Auer C. The crystalline
branes or choroidal new vessel membranes may have less lens after vitrectomy for diabetic retinopathy. Ophthalmol-
risk of postvitrectomy cataract regardless of age because ogy 1984;91:1480-4.
7. Margherio RR, Cox MS Jr, Trese MT, et al. Removal of
these macular conditions require relatively short surgery epimacular membranes. Ophthalmology 1985;92:1075-
and generally are less inflammatory?,11,14-l6 83.
We recognize the limitations of this series. First, the 8. Poliner LS, Olk RJ, Grand MG, et al. Surgical management of
retrospective nature of this report precluded the use of premacular fibroplasia. Arch Ophthalmol 1988; 106:761-4.
independent observers to eliminate bias in the grading of 9. Haimann MH, Abrams GW. Prevention of lens opacifica-
cataracts. Second, it was not possible to assess the impact tion during diabetic vitrectomy. Ophthalmology 1984;
91:116-21.
of postvitrectomy cataracts on visual outcome as patients 10. Maguire AM, Trese MT. Lens-sparing vitreoretinal surgery
underwent vitrectomy for a wide range of conditions and in infants. Arch Ophthalmol 1992; 110:284-6.
with a wide range of preoperative and postoperative visual 11. Melberg NS, Thomas MA. Nuclear sclerotic cataract after
function. Often the cause of postvitrectomy visual loss vitrectomy in patients young that 50 years of age. Ophthal-
was because of poor retinal function rather than dense mology 1995; 102:1466-71.
cataract. 12. Hakin KN, Lavin MJ, Leaver PK. Primary vitrectomy for
In our series, vitrectomy was performed for a wide rhegmatogenous retinal detachment. Graefes Arch Clin Exp
Ophthalmol 1993;231:344-6.
range of ocular conditions. Among these young patients, 13. Spencer WHo Ophthalmic Pathology. An Atlas and Text-
the rate of postvitrectomy cataract formation is compara- book, 3rd ed. Vol. 3. Philadelphia: W.B. Saunders Com-
ble to that of adult patients with similar conditions. The pany, 1986;2198-207.
risk of cataract after vitreous surgery appears to be related 14. Michels RG. Vitrectomy for macular pucker. Ophthalmol-
more to the complexity of the case than to the age of ogy 1984;91:1384-8.
the patient. This information may help guide vitreoretinal 15. Isernhagen RD, Smiddy WE, Michels RG, et al. Vitrectomy
for nondiabetic vitreous hemorrhage. Not associated with
surgeons in the planning of surgery and the counseling of vascular disease. Retina 1988;8:81-7.
young patients undergoing phakic pars plana vitrectomy. 16. Thomas MA, Dickinson JD, Melberg NS, et al. Visual re-
Acknowledgment. The authors thank Sarah M. Nusser, sults after surgical removal of subfoveal choroidal neovas-
PhD, for statistical assistance. cular membranes. Ophthalmology 1994; 101:1384-96.

1095