Material and methods

Peroperative anterior chamber aspirates obtained from 210 consecutive patients undergoing routine uncomplicated cataract extraction with posterior chamber intraocular lens (PCIOL) were examined. None had any active external eye or general infectious disease. Topical or oral antibiotics were not used pre- and peroperatively.

One hundred consecutive patients had a standard ECCE with PCIOL (group 1). One hundred and ten patients had phaco with PCIOL (group 2). One hour preoperatively the pupil was dilated with cyclopentolate 1% and phenylephrine 2.5%. The eye was anaesthetised using a peribulbar block with 10 ml mixture containing equal parts bupivacaine 0.5% and lignocaine 2% with 300–1500 IU hyaluronidase.

The skin of the eyelids and conjunctival sac was cleaned with povidone–iodine 10% and 5% respectively. A sterile drape with an adhesive foil was used to remove the eyelashes from the operative field. Irrigation was provided from a gravity feed bottle of Hartmann’s or balanced salt solution (BSS) containing 0.6 ml adrenaline 1/1000 per 500 ml. The infusion tubing did not contain a bacterial filter.

In group 1 a standard ECCE with PCIOL was performed using a limbal stepped section. The crystalline lens was then expressed and the remaining soft lens matter removed using a Simcoe irrigation/aspiration cannula. The syringes containing the AC aspirates were retained for culture. A PCIOL was then inserted in the posterior chamber and the posterior limbal wound was sutured with 10/0 nylon sutures and covered with conjunctiva.

In group 2 phaco was performed through a 3 mm scleral tunnel using a frown incision. A 1.5 mm corneal paracentesis was made for insertion of the second instrument. Continuous circular capsulorrhexis was completed under viscoelastic and the crystalline lens then extracted using the divide and conquer technique. The cortex was removed by automated irrigation/aspiration probe. The wound was extended to 5.5 mm and a one piece PMMA 5 × 6 mm intraocular lens (Iolab) inserted in the bag. The self sealing scleral wound was covered with conjunctiva. The phaco probe aspirates were poured into a sterile specimen bottle from the collection cassette and sent for culture. Fifty control specimen bottles containing sterile sodium chloride were sent among the study specimens.

MICROBIOLOGICAL STUDIES

The aspirates collected during ECCE and phaco were treated identically. A volume of 0.1 ml of fluid was inoculated onto three plates—5% blood agar, chocolate agar, and Sabouraud agar. The chocolate agar was incubated at 37°C in 5% carbon dioxide for 48 hours. The blood agar was incubated in an anaerobic cabinet for a total of 7 days and the Sabouraud plate at 37°C for 7 days aerobically.

In addition the fluid was centrifuged and the deposit resuspended in 2 ml of fluid. This was then inoculated into Pedi-bact blood culture broth (Organon-Technika) supplemented with 1 ml of Fildes extract. The bottles were incubated in the BacTalert machine for 7 days.

Any bottles flagging positive were unloaded and subcultured aerobically and anaerobically. Any positive growth, even if scanty, was recorded and isolates were identified to genes/ species level using standard methods.

STATISTICAL ANALYSIS

The χ² test was used to compare the results in the two groups. We calculated that our sample sizes would have an 80% chance of producing a statistically significant result at p<0.05 (epi info Stat Calc).

Results

In group 1 (ECCE), bacteria were cultured from AC aspirates in 29 cases (29%). There were two positive direct and 29 positive enrichment cultures (Table 1) which included; 32 coagulase negative staphylococcus (CNS), five Staphylococcus aureus, five diphtheroids, twoStreptococcus viridans, and one Moraxella.

In group 2 (phaco), bacteria were cultured from AC aspirates of 22 cases (20%). Six were from direct and 21 from enrichment cultures (Table 2). Enrichment cultures included; 15 CNS, two α haemolyticStreptococcus, two Acinetobacter species, one β haemolytic Streptococcus, one Moraxella, one Candida, and one non-fermentive oxidase positive Gram negative rod.

None of the 50 control bottles cultured positive. Although there was a lower rate of contamination of the AC aspirates during phaco, this difference between the two groups was not statistically significant (χ²=2.31: p=0.129).

Discussion

Despite technical improvements in the field of intraocular surgery and all the precautions taken, endophthalmitis is still a major postoperative complication. The reported incidence following ECCE is about 0.1%,5 ranging between 0.05% and 0.3%.6 However, the true figure is likely to be higher than this, many milder cases being both unrecognised and unreported and considered as sterile postoperative inflammation. The normal flora of the ocular surface and conjunctiva, especially CNS, is well known to produce postoperative inflammation and infections13 and is thought to be the main source of contamination. Irrigation of the conjunctiva before and during surgery introduces more micro-organisms from the conjunctival crypts onto the surface.7 Airborne microbes can also be delivered into the AC by means of intraocular lenses.2

It is accepted that preoperative use of povidone-iodine reduces the microbial contamination of the operating field significantly89 and was used in this study. Preoperative antibiotics were not used since they have only a limited effect on reducing ocular surface bacteria.10

Surgical techniques can alter the delivery rate of microbes into the AC and vitreous cavity. The incidence of endophthalmitis is higher following intracapsular cataract extraction5 when compared with ECCE. It is also higher in the event of a capsular rupture.11 Anterior chamber collapse and shallowing during the aspiration phase of ECCE can introduce microbes into the AC owing to a pressure lower than the atmosphere.2 This has been the basis for speculation that phaco with the use of a small incision and a better preservation of the AC throughout the surgery can reduce the rate of contamination.

Egger et al in two identical studies of 200 patients undergoing cataract extractions have shown less intraocular contamination with phaco.1213 The AC aspirates of 173 patients undergoing ECCE were compared with only 26 cases of phaco (Table 3). The study was rather weak because of the low numbers in the phaco group, although the results did reach statistical significance. Their conclusion was quoted in a subsequent editorial14and in support of a similar claim.

Manners et al compared two groups of 50 patients and reported 20% and 24% contamination during phaco and ECCE respectively. The difference was not statistically significant.15

We are not aware of larger studies to assess the effect of different surgical techniques on peroperative intraocular microbial contamination during cataract surgery. In this study the AC contamination during phaco (20%) was less than ECCE (29%), but this was not statistically significant (p>0.1). However, for a similar rate of positive cultures around 200 patients would be required in each group to achieve statistical significance (p<0.05). A meta-analysis on the data so far reported (Table 3) also supports the assertion that there is a reduced bacterial contamination of the AC during phaco (χ²=5.79: p<0.05).

Of the 58 positive cultures, 50 were obtained using enrichment techniques. There were only six positive direct cultures in phaco and two in the ECCE group (no statistical significance). Enrichment cultures reduce the chance of false negative cultures (Tables 1 and 2) and increase the likely detection of a low level of contamination. It is possible that cases with low levels of contamination produce a subclinical endophthalmitis and mild postoperative uveitis,24 though this requires further investigation.

Conclusion

Although open to question, to minimise the risk of postoperative endophthalmitis, the number of bacteria left in situ should be reduced to a minimum.16 Proper preoperative preparation of the eye and adherence to standard sterilisation protocols are important. The use of peroperative intracameral antibiotics has also been advocated.17

An uncomplicated, well controlled surgical technique is desirable. Small incision cataract extraction using phaco is ideally a controlled procedure that facilitates the maintenance of a well formed AC throughout the operation and is probably associated with less peroperative bacterial influx. In this study we have been unable to prove that the lower rate of bacterial contamination of the AC during phaco was not due to chance alone.

We have found enrichment cultures valuable with a much higher sensitivity to identify bacteria. Their use, especially for detection of low grade infection, seems to minimise the rate of false negative results.