Downloaded from http://bjo.bmj.com/ on October 22, 2014 - Published by group.bmj.com
BJO Online First, published on October 17, 2014 as 10.1136/bjophthalmol-2014-305028 Clinical science
Autologous blood versus fibrin glue for conjunctival autograft adherence in sutureless pterygium surgery: a randomised controlled trial Abraham Kurian, Iodine Reghunadhan, K G R Nair ▸ Additional material is published online only. To view please visit the journal online (http://dx.doi.org/10.1136/ bjophthalmol-2014-305028). Chaithanya Eye Hospital and Research Institute, Trivandrum, Kerala, India Correspondence to Dr Abraham Kurian, Chaithanya Eye Hospital and Research Institute, Trivandrum, Kerala 695004, India; [email protected] Received 31 January 2014 Revised 14 July 2014 Accepted 21 September 2014
To cite: Kurian A, Reghunadhan I, Nair KGR. Br J Ophthalmol Published Online First: [ please include Day Month Year] doi:10.1136/bjophthalmol2014-305028
ABSTRACT Aims To assess the efficacy of securing conjunctival autograft (CAG) without glue or sutures, using the patient’s own blood at the surgical site, and to compare it with the current accepted standard of using fibrin glue for graft adherence, in pterygium surgery. Methods A single-centre, prospective, randomised controlled trial was carried out in 200 eyes of patients with primary pterygia who were advised pterygium excision with CAG. Enrolled participants were assigned to Group I or II by randomisation. After excision of pterygium, they underwent CAG with autologous blood in Group I (100 eyes) and CAG with fibrin glue in Group II (100 eyes). During follow-up of 1 year, the eyes were assessed for graft adherence and recurrence. Results Of the 200 eyes randomised, 6 eyes that did not complete intended follow-up were excluded from final analysis. Of the 194 eyes (Group I n=96, Group II n=98), on the first postoperative day, 3 eyes in Group I (3.13%) had total graft dislodgement requiring regrafting from another site or reattachment with glue. In Group II also 2 eyes (2.04%) had graft dislodgement on the first postoperative day requiring regrafting from another site. During the 1-year follow-up, 6 eyes in Group I (6.25%) and 8 eyes in Group II (8.16%) developed recurrence. Conclusions Feasibilty of adherence of the graft without glue in pterygium surgery is promising and has results comparable with the fibrin glue technique in terms of long-term outcome and recurrence, suggesting the potential for autologous blood to replace fibrin glue in graft fixation. Trial registration number Clinical Trial Registry, India: CTRI/2013/06/003764 and UTN: U1111-1140-6572.
above criteria and gained popularity in the 1980s following the landmark article by Kenyon et al in 19855 who reported a low recurrence rate of 5.3%. Prabhasawat et al (1997)6 reported that autologous CAG is the best method, with a low recurrence rate and higher safety margin. Thereafter, several prospective randomised trials of CAG for pterygium surgery have reported higher recurrence rates of 16%–39% in high-risk population.7 8 Nonetheless, CAG has now been in vogue for more than 3 decades and has stood the test of time as the standard of care in pterygium surgery. Traditionally, Pterygium surgery with CAG used sutures to secure the autograft in place. The use of fibrin glue for this purpose was popularised by Koranyi et al in 20049 10 who used Tisseel, a twocomponent tissue adhesive which mimics the natural fibrin clot formation (figure 1). Studies comparing fibrin glue and sutures have found that the grafts secured with fibrin glue are as stable with significantly less inflammation and, consequently, less recurrence, compared to sutures.11 12 Hence, free CAG with fibrin glue can authentically be stated as being the preferred choice for pterygium surgery today. Our study intended to evaluate the technique of securing the CAG with fibrin, using the normal physiological clot from the patient’s own blood at the surgical site acting as a bioadhesive, and to compare it with the current standard of securing it with commercially available fibrin sealant, with regard to graft adherence and recurrence. A noninferiority trial with a margin (δ) defined at 5% was planned to establish that the new intervention was at least as good as the current standard.
INTRODUCTION
MATERIALS AND METHODS Patients
Historically, pterygium and its management was first described in 1000 BC by Susruta.1 2 Thereafter, though innumerable surgical options for pterygium have emerged and evolved through three millennia, complications like recurrence after surgery have been vexing. The bare sclera technique which was popular from 1960 to the early 1980s has been abandoned due to the unacceptable recurrence rate, often as high as 60%–80%.3 Jaros and DeLuise4 reported recurrence rate of 24%–89% with the bare sclera technique, while use of adjunctive chemotherapeutic agents have been viewed with caution due to the chance for potentially sight-threatening complications. The logical approach is a safe method with low recurrence rate and free of long-term complications. Conjunctival autograft (CAG) fulfils the
Patients, 18–80 years of age, with primary pterygia, who were advised surgery were considered for enrolment. Exclusion criteria: ▸ previous ocular surgery; ▸ significant ocular surface disease; ▸ known hypersensitivity to any component of fibrin glue; ▸ patients on anticoagulants or antiplatelet agents; ▸ known bleeding or coagulation disorder.
Design A single-centre, prospective, randomised, parallel group, double blind, non-inferiority trial. The trial conformed to the tenets of the Declaration of Helsinki and was registered in the
et al. Br J Ophthalmol doi:10.1136/bjophthalmol-2014-305028 Copyright Article author (or Kurian theirA,employer) 2014.2014;0:1–7. Produced by BMJ Publishing Group Ltd under licence.
1
Downloaded from http://bjo.bmj.com/ on October 22, 2014 - Published by group.bmj.com
Clinical science
Figure 1 2014 Baxter Healthcare Corporation. (A) Normal physiological fibrin clot. Scanning electron microscope image: 6000×. (B) Fibrin clot formed with commercially available fibrin sealant which mimics the natural clot. Clinical Trial Registry, India (ctri.nic.in) CTRI/2013/06/003764, UTN: U1111-1140-6572. Ethics Committee approval was attained (Ethical Committee, Chaithanya Eye Hospital and Research Institute, Trivandrum) and informed consent obtained from all participants.
Sample size estimation Sample size was determined as for a binary outcome noninferiority trial13 using graft adherence as the primary outcome measure, assuming 98% success rates in both groups and defining the non-inferiority limit (δ) as 5%. With a 5% significance level and power of 80%, a sample size of 97 patients in each group was necessary. An inclusion period of 12 months was anticipated to achieve this. The study took place at the Chaithanya Eye Hospital and Research Institute, Trivandrum, a tertiary eye care centre in Kerala, South India, from January 2012 to December 2013. Enrolled patients were randomly assigned into two groups— Group I (100 eyes) underwent CAG with autologous blood, wherein graft fixation was spontaneous using the physiological fibrin clot formed at the surgical site. Group II (100 eyes) underwent CAG with fibrin glue (Tisseel, Baxter International Inc., Illinois, USA). To eliminate bias, patients enrolled into the study, when they reported for the surgery, were randomly assigned into one of the groups by the personnel in the coordinating centre using a computer generated random number table using permuted fixed blocked randomisation in blocks of 10 patients to ensure equal number of persons in each arm. The group assignment was masked to the participants and assigners, using sequentially numbered, sealed, opaque envelopes. Also to reduce intraobserver bias and to minimise the influence of the known surgical technique on the extent of dissection and size of graft, group assignment was blinded to the surgeon till the graft was harvested. Postoperatively, follow-up was done by a single observer who was masked to the group assignment and the intervention undergone.
Surgical technique All surgeries were performed after routine blood investigations. Under local infiltrative anaesthesia using 2% lignocaine; in both groups, pterygia were excised by avulsing the head with conjunctival forceps. Subsequent dissection was done with Westcott 2
scissors and cornea cleaned of remnants. The recipient bed was measured with surgical callipers, and a graft size at least 1 mm larger was aimed at. The average graft size was 6×5 mm in all the cases. From the superotemporal quadrant, a thin CAG was dissected free of Tenon’s and without buttonholing. The graft was then pulled onto the cornea taking care to prevent it from getting rolled over, to maintain epithelial side up orientation. In Group I eyes, bleeders were not cauterised at all or only minimal cautery, if any, was preferred. After excision of pterygium, blood was allowed to collect in the bare recipient scleral bed for 1–2 min. The free CAG was applied over the bare recipient bed, spread and held in position for 3 min until firm setting of the autologous fibrin occurred in most cases, ensuring graft adhesion. While waiting for the adhesion, care was taken to ensure that excessive and prolonged bleeding did not displace the graft from the bed, and residual active haemorrhages beneath the graft, if any, were tamponaded by direct compression with a blunt instrument. Graft adherence was confirmed at the end of the procedure, antibiotic eye ointment applied and a pressure bandage was given for 24 h. In Group II eyes, bleeders were cauterised as required, the two components of fibrin sealant (Tisseel, Baxter) reconstituted according to the manufacturer’s guidelines were loaded separately in two separate syringes. One drop from each was placed over the recipient bed and the prepared graft was pulled onto it and smoothened out. Uniform graft adhesion was confirmed, antibiotic eye ointment was applied and a simple pad and bandage was given for 24 h. Postoperatively, both groups of patients were followed up at Day 1, Week 1, Month 1, Month 3, Month 6 and 1 year, postoperatively. Routine postoperative therapy was initiated in both groups. This consisted of 0.5% Moxifoxacin eye drops every 6 h for 1 week; 1% prednisolone acetate eye drops every 4 h for 1 week to be tapered over a month and lubricant eye drops every 4 h for 1 month. At each visit, a slit lamp examination was performed to assess the graft adherence, displacement or retraction and presence of recurrence or any other untoward effect.
Main outcome measures The primary outcomes assessed were graft adherence at 1 day and 1 week, postoperatively, and recurrence at the end of 1 year. Kurian A, et al. Br J Ophthalmol 2014;0:1–7. doi:10.1136/bjophthalmol-2014-305028
Downloaded from http://bjo.bmj.com/ on October 22, 2014 - Published by group.bmj.com
Clinical science Recurrence was defined as any postoperative regrowth of fibrovascular tissue crossing the corneoscleral limbus into the clear cornea. Secondary outcomes studied included graft-related complications, like partial graft retraction, subconjunctival haemorrhage and granuloma formation. Partial graft retraction was defined as retraction of the graft from either the conjunctival or limbal side for at least 1 mm.
Statistical methods The primary outcomes were compared between the two groups using the z test for difference in proportions adapted for noninferiority hypothesis.14 15 The hypothesis testing used a noninferiority margin (δ) of 5%. Non-inferiority was concluded at level α (0.05) if z>z1-α. (z1-α =1.64), equivalent to a one-tailed p value −δ rejected the null hypothesis. Incidence of partial graft retraction between the groups was also analysed by the z test for difference in proportions. The incidence of subconjuntival haemorrhage postoperatively between the two groups was analysed for statistical significance by Pearson’s χ2 test.
RESULTS Figure 2 shows the number of patients and dropouts at each stage. So, the final analysis had 194 eyes—Group I n=96 and group II n=98.
Table 1
Patient demographics in Group I and Group II
Variable
Group I
Group II
Age (mean±SD) years Sex (number of patients): Male Female Total Site of pterygium (number of eyes): Nasal Temporal
42.5±10.4
37.4±12.6
50 46 96
58 40 98
93 3
92 6
Table 1 summarises the patient demographics and variables in the two groups which were comparable.
Main outcome measures Table 2 summarises the main outcomes and their comparison between the two groups. Graft adherence: Graft adherence was assessed by the total dislodgement of the graft requiring intervention. On the first postoperative day, all the grafts showed good graft adherence in Group I, except 3 eyes which had graft dislodgement, of which 2 eyes required reattachment with glue and the other required regrafting from another site and fixation with glue due to loss of the graft. In Group II, 2 eyes had total graft dislodgement on the 1st postoperative day with lost graft
Figure 2 Consort 2010 flow diagram.
Kurian A, et al. Br J Ophthalmol 2014;0:1–7. doi:10.1136/bjophthalmol-2014-305028
3
Downloaded from http://bjo.bmj.com/ on October 22, 2014 - Published by group.bmj.com
Clinical science Table 2 Comparison of main outcomes in both groups Group I Complication (n=96)
z-statistics p Value Group II (n=98) (non-inferiority) (one-tailed)
Total graft dislodgement % success Recurrence (1 year) % success
3 (3.13%)
2 (2.04%)
96.87 6 (6.25%)
97.96 8 (8.16%)
1.719
0.0428
93.75
91.84
1.863
0.0312
Table 3 Incidence of partial graft retraction. Comparison between the two groups at 1week
Number of cases (%) % success z-Statisitics (non-inferiority) p-Value (one-tailed)
Group I (n=96)
Group II (n=98)
10 (10.4) 89.6 0.6628 0.2537
8 (8.16) 91.84
Secondary outcomes which, in turn, required regrafting from another site in both eyes. The rest of the eyes showed good apposition of graft on the first day which was maintained at 1 week (figure 3). From table 2, regarding graft dislodgement, the difference in success rate between the test group (Group I, pT) and the control group (Group II, pC) was −1.09% (CI −4.84% to 2.66%). The LB −4.84% being > −δ (−5%) rejected the null hypothesis and proved non-inferiority. The Statistical analysis using the z test for difference in proportions for non-inferiority hypothesis also resulted in a significant p value −δ) in favour of noninferiority. Comparison by z test for difference in proportions between the groups using non-inferiority hypothesis also yielded a p
BJO Online First, published on October 17, 2014 as 10.1136/bjophthalmol-2014-305028 Clinical science
Autologous blood versus fibrin glue for conjunctival autograft adherence in sutureless pterygium surgery: a randomised controlled trial Abraham Kurian, Iodine Reghunadhan, K G R Nair ▸ Additional material is published online only. To view please visit the journal online (http://dx.doi.org/10.1136/ bjophthalmol-2014-305028). Chaithanya Eye Hospital and Research Institute, Trivandrum, Kerala, India Correspondence to Dr Abraham Kurian, Chaithanya Eye Hospital and Research Institute, Trivandrum, Kerala 695004, India; [email protected] Received 31 January 2014 Revised 14 July 2014 Accepted 21 September 2014
To cite: Kurian A, Reghunadhan I, Nair KGR. Br J Ophthalmol Published Online First: [ please include Day Month Year] doi:10.1136/bjophthalmol2014-305028
ABSTRACT Aims To assess the efficacy of securing conjunctival autograft (CAG) without glue or sutures, using the patient’s own blood at the surgical site, and to compare it with the current accepted standard of using fibrin glue for graft adherence, in pterygium surgery. Methods A single-centre, prospective, randomised controlled trial was carried out in 200 eyes of patients with primary pterygia who were advised pterygium excision with CAG. Enrolled participants were assigned to Group I or II by randomisation. After excision of pterygium, they underwent CAG with autologous blood in Group I (100 eyes) and CAG with fibrin glue in Group II (100 eyes). During follow-up of 1 year, the eyes were assessed for graft adherence and recurrence. Results Of the 200 eyes randomised, 6 eyes that did not complete intended follow-up were excluded from final analysis. Of the 194 eyes (Group I n=96, Group II n=98), on the first postoperative day, 3 eyes in Group I (3.13%) had total graft dislodgement requiring regrafting from another site or reattachment with glue. In Group II also 2 eyes (2.04%) had graft dislodgement on the first postoperative day requiring regrafting from another site. During the 1-year follow-up, 6 eyes in Group I (6.25%) and 8 eyes in Group II (8.16%) developed recurrence. Conclusions Feasibilty of adherence of the graft without glue in pterygium surgery is promising and has results comparable with the fibrin glue technique in terms of long-term outcome and recurrence, suggesting the potential for autologous blood to replace fibrin glue in graft fixation. Trial registration number Clinical Trial Registry, India: CTRI/2013/06/003764 and UTN: U1111-1140-6572.
above criteria and gained popularity in the 1980s following the landmark article by Kenyon et al in 19855 who reported a low recurrence rate of 5.3%. Prabhasawat et al (1997)6 reported that autologous CAG is the best method, with a low recurrence rate and higher safety margin. Thereafter, several prospective randomised trials of CAG for pterygium surgery have reported higher recurrence rates of 16%–39% in high-risk population.7 8 Nonetheless, CAG has now been in vogue for more than 3 decades and has stood the test of time as the standard of care in pterygium surgery. Traditionally, Pterygium surgery with CAG used sutures to secure the autograft in place. The use of fibrin glue for this purpose was popularised by Koranyi et al in 20049 10 who used Tisseel, a twocomponent tissue adhesive which mimics the natural fibrin clot formation (figure 1). Studies comparing fibrin glue and sutures have found that the grafts secured with fibrin glue are as stable with significantly less inflammation and, consequently, less recurrence, compared to sutures.11 12 Hence, free CAG with fibrin glue can authentically be stated as being the preferred choice for pterygium surgery today. Our study intended to evaluate the technique of securing the CAG with fibrin, using the normal physiological clot from the patient’s own blood at the surgical site acting as a bioadhesive, and to compare it with the current standard of securing it with commercially available fibrin sealant, with regard to graft adherence and recurrence. A noninferiority trial with a margin (δ) defined at 5% was planned to establish that the new intervention was at least as good as the current standard.
INTRODUCTION
MATERIALS AND METHODS Patients
Historically, pterygium and its management was first described in 1000 BC by Susruta.1 2 Thereafter, though innumerable surgical options for pterygium have emerged and evolved through three millennia, complications like recurrence after surgery have been vexing. The bare sclera technique which was popular from 1960 to the early 1980s has been abandoned due to the unacceptable recurrence rate, often as high as 60%–80%.3 Jaros and DeLuise4 reported recurrence rate of 24%–89% with the bare sclera technique, while use of adjunctive chemotherapeutic agents have been viewed with caution due to the chance for potentially sight-threatening complications. The logical approach is a safe method with low recurrence rate and free of long-term complications. Conjunctival autograft (CAG) fulfils the
Patients, 18–80 years of age, with primary pterygia, who were advised surgery were considered for enrolment. Exclusion criteria: ▸ previous ocular surgery; ▸ significant ocular surface disease; ▸ known hypersensitivity to any component of fibrin glue; ▸ patients on anticoagulants or antiplatelet agents; ▸ known bleeding or coagulation disorder.
Design A single-centre, prospective, randomised, parallel group, double blind, non-inferiority trial. The trial conformed to the tenets of the Declaration of Helsinki and was registered in the
et al. Br J Ophthalmol doi:10.1136/bjophthalmol-2014-305028 Copyright Article author (or Kurian theirA,employer) 2014.2014;0:1–7. Produced by BMJ Publishing Group Ltd under licence.
1
Downloaded from http://bjo.bmj.com/ on October 22, 2014 - Published by group.bmj.com
Clinical science
Figure 1 2014 Baxter Healthcare Corporation. (A) Normal physiological fibrin clot. Scanning electron microscope image: 6000×. (B) Fibrin clot formed with commercially available fibrin sealant which mimics the natural clot. Clinical Trial Registry, India (ctri.nic.in) CTRI/2013/06/003764, UTN: U1111-1140-6572. Ethics Committee approval was attained (Ethical Committee, Chaithanya Eye Hospital and Research Institute, Trivandrum) and informed consent obtained from all participants.
Sample size estimation Sample size was determined as for a binary outcome noninferiority trial13 using graft adherence as the primary outcome measure, assuming 98% success rates in both groups and defining the non-inferiority limit (δ) as 5%. With a 5% significance level and power of 80%, a sample size of 97 patients in each group was necessary. An inclusion period of 12 months was anticipated to achieve this. The study took place at the Chaithanya Eye Hospital and Research Institute, Trivandrum, a tertiary eye care centre in Kerala, South India, from January 2012 to December 2013. Enrolled patients were randomly assigned into two groups— Group I (100 eyes) underwent CAG with autologous blood, wherein graft fixation was spontaneous using the physiological fibrin clot formed at the surgical site. Group II (100 eyes) underwent CAG with fibrin glue (Tisseel, Baxter International Inc., Illinois, USA). To eliminate bias, patients enrolled into the study, when they reported for the surgery, were randomly assigned into one of the groups by the personnel in the coordinating centre using a computer generated random number table using permuted fixed blocked randomisation in blocks of 10 patients to ensure equal number of persons in each arm. The group assignment was masked to the participants and assigners, using sequentially numbered, sealed, opaque envelopes. Also to reduce intraobserver bias and to minimise the influence of the known surgical technique on the extent of dissection and size of graft, group assignment was blinded to the surgeon till the graft was harvested. Postoperatively, follow-up was done by a single observer who was masked to the group assignment and the intervention undergone.
Surgical technique All surgeries were performed after routine blood investigations. Under local infiltrative anaesthesia using 2% lignocaine; in both groups, pterygia were excised by avulsing the head with conjunctival forceps. Subsequent dissection was done with Westcott 2
scissors and cornea cleaned of remnants. The recipient bed was measured with surgical callipers, and a graft size at least 1 mm larger was aimed at. The average graft size was 6×5 mm in all the cases. From the superotemporal quadrant, a thin CAG was dissected free of Tenon’s and without buttonholing. The graft was then pulled onto the cornea taking care to prevent it from getting rolled over, to maintain epithelial side up orientation. In Group I eyes, bleeders were not cauterised at all or only minimal cautery, if any, was preferred. After excision of pterygium, blood was allowed to collect in the bare recipient scleral bed for 1–2 min. The free CAG was applied over the bare recipient bed, spread and held in position for 3 min until firm setting of the autologous fibrin occurred in most cases, ensuring graft adhesion. While waiting for the adhesion, care was taken to ensure that excessive and prolonged bleeding did not displace the graft from the bed, and residual active haemorrhages beneath the graft, if any, were tamponaded by direct compression with a blunt instrument. Graft adherence was confirmed at the end of the procedure, antibiotic eye ointment applied and a pressure bandage was given for 24 h. In Group II eyes, bleeders were cauterised as required, the two components of fibrin sealant (Tisseel, Baxter) reconstituted according to the manufacturer’s guidelines were loaded separately in two separate syringes. One drop from each was placed over the recipient bed and the prepared graft was pulled onto it and smoothened out. Uniform graft adhesion was confirmed, antibiotic eye ointment was applied and a simple pad and bandage was given for 24 h. Postoperatively, both groups of patients were followed up at Day 1, Week 1, Month 1, Month 3, Month 6 and 1 year, postoperatively. Routine postoperative therapy was initiated in both groups. This consisted of 0.5% Moxifoxacin eye drops every 6 h for 1 week; 1% prednisolone acetate eye drops every 4 h for 1 week to be tapered over a month and lubricant eye drops every 4 h for 1 month. At each visit, a slit lamp examination was performed to assess the graft adherence, displacement or retraction and presence of recurrence or any other untoward effect.
Main outcome measures The primary outcomes assessed were graft adherence at 1 day and 1 week, postoperatively, and recurrence at the end of 1 year. Kurian A, et al. Br J Ophthalmol 2014;0:1–7. doi:10.1136/bjophthalmol-2014-305028
Downloaded from http://bjo.bmj.com/ on October 22, 2014 - Published by group.bmj.com
Clinical science Recurrence was defined as any postoperative regrowth of fibrovascular tissue crossing the corneoscleral limbus into the clear cornea. Secondary outcomes studied included graft-related complications, like partial graft retraction, subconjunctival haemorrhage and granuloma formation. Partial graft retraction was defined as retraction of the graft from either the conjunctival or limbal side for at least 1 mm.
Statistical methods The primary outcomes were compared between the two groups using the z test for difference in proportions adapted for noninferiority hypothesis.14 15 The hypothesis testing used a noninferiority margin (δ) of 5%. Non-inferiority was concluded at level α (0.05) if z>z1-α. (z1-α =1.64), equivalent to a one-tailed p value −δ rejected the null hypothesis. Incidence of partial graft retraction between the groups was also analysed by the z test for difference in proportions. The incidence of subconjuntival haemorrhage postoperatively between the two groups was analysed for statistical significance by Pearson’s χ2 test.
RESULTS Figure 2 shows the number of patients and dropouts at each stage. So, the final analysis had 194 eyes—Group I n=96 and group II n=98.
Table 1
Patient demographics in Group I and Group II
Variable
Group I
Group II
Age (mean±SD) years Sex (number of patients): Male Female Total Site of pterygium (number of eyes): Nasal Temporal
42.5±10.4
37.4±12.6
50 46 96
58 40 98
93 3
92 6
Table 1 summarises the patient demographics and variables in the two groups which were comparable.
Main outcome measures Table 2 summarises the main outcomes and their comparison between the two groups. Graft adherence: Graft adherence was assessed by the total dislodgement of the graft requiring intervention. On the first postoperative day, all the grafts showed good graft adherence in Group I, except 3 eyes which had graft dislodgement, of which 2 eyes required reattachment with glue and the other required regrafting from another site and fixation with glue due to loss of the graft. In Group II, 2 eyes had total graft dislodgement on the 1st postoperative day with lost graft
Figure 2 Consort 2010 flow diagram.
Kurian A, et al. Br J Ophthalmol 2014;0:1–7. doi:10.1136/bjophthalmol-2014-305028
3
Downloaded from http://bjo.bmj.com/ on October 22, 2014 - Published by group.bmj.com
Clinical science Table 2 Comparison of main outcomes in both groups Group I Complication (n=96)
z-statistics p Value Group II (n=98) (non-inferiority) (one-tailed)
Total graft dislodgement % success Recurrence (1 year) % success
3 (3.13%)
2 (2.04%)
96.87 6 (6.25%)
97.96 8 (8.16%)
1.719
0.0428
93.75
91.84
1.863
0.0312
Table 3 Incidence of partial graft retraction. Comparison between the two groups at 1week
Number of cases (%) % success z-Statisitics (non-inferiority) p-Value (one-tailed)
Group I (n=96)
Group II (n=98)
10 (10.4) 89.6 0.6628 0.2537
8 (8.16) 91.84
Secondary outcomes which, in turn, required regrafting from another site in both eyes. The rest of the eyes showed good apposition of graft on the first day which was maintained at 1 week (figure 3). From table 2, regarding graft dislodgement, the difference in success rate between the test group (Group I, pT) and the control group (Group II, pC) was −1.09% (CI −4.84% to 2.66%). The LB −4.84% being > −δ (−5%) rejected the null hypothesis and proved non-inferiority. The Statistical analysis using the z test for difference in proportions for non-inferiority hypothesis also resulted in a significant p value −δ) in favour of noninferiority. Comparison by z test for difference in proportions between the groups using non-inferiority hypothesis also yielded a p
