Effect of anterior chamber air tamponade after phacoemulsification cataract surgery: a systematic review and meta-analysis

Abstract

Background

This study aimed to evaluate the safety of anterior chamber air tamponade post-phacoemulsification on the cornea compared to balanced salt solution or Ringer’s lactate filling, and summarize its potential influence on the aspects like the risk of postoperative endophthalmitis and Descemet’s membrane detachment (DMD).

Methods

This meta-analysis was conducted using Review Manager 5.4 and included studies that evaluated post-phacoemulsification air tamponade compared with no air tamponade before October 2025. Methodological quality was assessed using the Cochrane Risk of Bias tool. This review was registered in PROSPERO (CRD420251177446) and followed PRISMA guidelines.

Results

Among seven included studies (1340 eyes), there were no significant between-group differences (mean difference [95% confidence interval]) at 1 month, with or without air tamponade, in the endothelial cell density (−103.59 [−256.86 to 49.68] cells/mm²; p = 0.19), central corneal thickness (−4.86 [−19.81 to 10.08] µm; p = 0.52), or coefficient of variation of the corneal endothelial cell size (−0.01 [−0.04 to 0.01]; p = 0.22). A minimal yet significant difference in hexagonality favored the no-air-tamponade group (−0.01 [−0.03 to −0.00]; p = 0.03).

Conclusions:

Intracameral air injection demonstrated comparable short-term corneal endothelial safety, with the potential added benefits of reduced inflammation and infection risk and of protection of the Descemet’s membrane.

Trial registration

PROSPERO, CRD420251177446. Registered 2025.

Clinical trial number

Not applicable.

Background

Cataract is the primary cause of reversible vision loss [1], with approximately 95 million people affected worldwide [2]. Considering its general good outcome in visual rehabilitation with a low complication rate, phacoemulsification has become the gold standard for cataract extraction in most countries [1, 2]. After the phacoemulsification, the clear corneal incision should be well-sealed while maintaining the depth of anterior chamber (AC) [3–7]. This enables the maintenance of physiological intraocular pressure (IOP), prevents the inflow of ocular surface fluid into the AC, and stabilizes the intraocular lens (IOL) position [8–11]. Air tamponade has been proposed as a technique with acceptable safety that may offer certain advantages over conventional BSS/Ringer’s lactate reformation for sealing clear corneal incisions, but its potential benefits and risks remain incompletely defined. For example, air tamponade has been suggested to mitigate phacoemulsification-induced DMD [12, 13], and to potentially reduce endophthalmitis risk through physical barrier effects [14]. However, these putative benefits need to be validated by a larger sample of research; furthermore, concerns exist regarding potential endothelial effects.

Air tamponade has the most prominent effect on the corneas, whereas phacoemulsification has proven deleterious to the corneal endothelium [15–17]. This endothelium comprises a single layer of endothelial cells situated between AC and corneal stroma and forms a barrier against the aqueous humor [18, 19]. It is essential for maintaining corneal stromal hydration at approximately 78% and corneal transparency that functions as a bicarbonate “pump” [20–22]. The corneal endothelium has limited regenerative ability; therefore, decreased density or function that exceeds its critical threshold can cause permanent endothelial functional decompensation [23, 24]. During cataract surgery, the energy of phacoemulsification, mechanical trauma, increased IOP, and AC surging could all injure the corneal endothelium, and thereby decrease the endothelial cell count and impair cell structure and function. Patients may experience corneal edema and even irreversible bullous keratopathy, which necessitates corneal tissue transplantation [25–28].

Under these circumstances, air injection may help restore normal IOP and AC, isolating the corneal endothelial layer from the water and thus protecting the corneal endothelium from surgical events. However, direct contact between sterilized air and the corneal endothelium is unnatural, and it is unclear whether such contact confers any potential detrimental effects. Additionally, current literature on this issue presents contradictory findings. Therefore, we performed a systematic review to comprehensively evaluate effects of anterior chamber air tamponade post-phacoemulsification on corneal endothelial parameters, IOL positional stability, and risks of postoperative endophthalmitis and DMD. Furthermore, by performing a meta-analysis of available clinical studies, we sought to clarify whether air tamponade following phacoemulsification confers superior corneal endothelium protection.

Methods

Eligibility criteria

All studies that compared effects on the cornea with air injection without air injection into the AC at the end of phacoemulsification were included. The group receiving intracameral air injection for anterior chamber reformation at the conclusion of phacoemulsification was defined as the intervention group; the group receiving balanced salt solution (BSS) or Ringer’s lactate injection was defined as the control. The review was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO 2025; registration number: CRD420251177446).

The inclusion criteria were as follows: clinical studies, including randomized controlled trials (RCTs) and retrospective studies, and comparative studies, that compared air tamponade with irrigation fluid injected into the AC at the conclusion of phacoemulsification. There were no restrictions regarding age, sex, or disease status. No language restrictions were imposed. Studies that were available only in abstract form, review articles, case reports, and single-arm observational studies that lacked a control group were excluded.

Study selection

This systematic review and meta-analysis were conducted in compliance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Two independent reviewers were responsible for screening the titles and abstracts of the retrieved records. For studies that were deemed relevant, full-text articles were retrieved and then assessed for eligibility according to the predefined inclusion criteria.

Search methods for identifying studies

Electronic databases, including PubMed, Web of Science, Embase, Scopus, Cochrane Library, and China National Knowledge Infrastructure, were searched by the author. The final search was performed on October 27, 2025. The thesaurus headings, search operators, and restrictions were adjusted for each database. The search terminologies included “Air,” “Air Injection,” “Air tamponade,” “Air bubble,” “Cataract Surgery,” and “Phacoemulsification.”

Data collection and risk of bias assessment

Relevant data were extracted from included studies by two independent reviewers using a standardized electronic form. Any discrepancies were addressed through discussion with a third reviewer. The category of collected information included study characteristics, participant demographics, surgical details, intervention protocols, and outcomes of interest.

The quality and risk of bias of the included articles were assessed by at least two independent reviewers. The risk of bias in RCTs was identified using the Cochrane Risk of Bias, Version 2 (RoB-2) tool; the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) tool was used to evaluate non-RCTs. All the included studies were classified as having a low, unclear, or high risk of bias.

Data synthesis and analysis

Outcomes reported in at least two studies were included in the meta-analysis. The mean difference (MD) was applied for evaluating continuous variables between the intervention and control groups. Review Manager 5.4 was used for the data analysis. Both fixed- and random-effect models were used. The results are presented as forest plots with 95% confidence intervals. Cochran’s Q test (chi-square) was used to evaluate heterogeneity and the I² statistic was used to measure inconsistency.

Results

Studies included in the meta-analysis

An overview of the study-selection process is presented in the PRISMA flow diagram in Fig. 1. In total, 1067 results were identified, and 342 articles underwent title/abstract screening after deduplication. Next, 23 articles underwent a full-text review. Finally, seven articles were included in the meta-analysis.

Fig. 1.

PRISMA flow diagram illustrating the study selection process for the systematic review and meta-analysis comparing air tamponade and stromal hydration in cataract surgery. PRISMA, Preferred Reporting Items for systematic Reviews and meta-Analyses

Study characteristics of studies included in the meta-analysis

The articles included in this meta-analysis were predominantly prospective (n = 4), retrospective (n = 2), and case–control studies (n = 1) [29–35]. These studies encompassed diverse populations from multiple regions and countries, including China, Egypt, Greece, Turkey, Iraq, and Taiwan, which indicated that their conclusions have broad applicability.

Patient characteristics of studies included in the meta-analysis

The seven studies collectively comprised 1340 eyes, with the patient age ranging from 50 to 85 years. Although the preoperative endothelial cell density (ECD), central corneal thickness (CCT), coefficient of variation (CV), and hexagonality (HEX) values varied slightly across the studies, they were well-balanced between the groups (Table 1).

Table 1.

Comparison of patient demographics and preoperative and operative information in the included studies

Study source	Group	Sample size (eyes)	Age (Years)	Incision size (mm)	Sex (Female)	Cataract hardness	Phacoemulsification energy (J)	Preoperative ECD (cells/mm2)	Preoperative CCT (μm)	Preoperative CV (%)	Preoperative HEX (%)
Keles 2022 [33]	Air Bubble Group	71	65.90 ± 11.05	2.75	25	Grade 2	Not Provided	2427.01 ± 31.95	526.08 ± 31.69	40.46 ± 5.12	44.92 ± 6.80
	No Air Bubble Group	63	65.02 ± 10.38	2.75	20	Grade 2	Not Provided	2453.24 ± 278.52	532.54 ± 35.29	40.94 ± 7.38	44.32 ± 5.98
Alsmman 2018 [29]	Air Group	250	57.90 ± 3.90	2.75	150	Grade 2	6.51 ± 1.0	2604 ± 367	509 ± 34	40.30 ± 8.40	51.00 ± 10.00
	BSS Group	250	56.00 ± 2.30	2.75	130	Grade 2	7.19 ± 1.1	2646 ± 284	508 ± 22	38.20 ± 6.00	49.00 ± 8.00
Zaki 2021 [31]	Air Group	38	62.18 ± 10.51	Not Provided	10	Grade 1–2	23.5 ± 7.3	2652.2 ± 371.2	494.26 ± 31.69	38.04 ± 6.5	55.50 ± 10.07
	Ringer’s Lactate Group	38	65.05 ± 5.72	Not Provided	12	Grade 1–2	25.2 ± 9.1	2557.8 ± 454.7	510.68 ± 36.39	35.7 ± 5.2	54.10 ± 13.40
Chen 2022 [32]	IABT Group	201	67.90 ± 11.96	2.65	118	Not Provided	25.36 ± 21.02	2537 ± 430	549.80 ± 35.20	33.85 ± 10.73	64.18 ± 10.07
	Non-IABT Group	233	69.12 ± 11.20	2.65	117	Not Provided	25.41 ± 20.72	2583 ± 391	554.60 ± 34.90	33.42 ± 11.41	64.67 ± 8.74
Yuksel 2019 [34]	Air Bubble Injection Group	20	Not Provided	2.8	(Total: 18/10)	Grade 3	7.6 ± 3.4	2091.0 ± 80.9	Not Provided	Not Provided	Not Provided
	Without Air Bubble Injection Group	20	Not Provided	2.8	(Total: 18/10)	Grade 3	7.9 ± 3.8	2096.5 ± 90.9	Not Provided	Not Provided	Not Provided
Matafisi 2020 [30]	Air bubble Group	49	72 (66, 76)	2.4	27	Not Provided	8.17 (5.03–13.7)	2428.41 ± 355.01	535.94 ± 29.84	Not Provided	Not Provided
	Hydration Group	49	71 (64, 78)	2.4	20	Not Provided	7.57 (4.99–13.42)	2505.24 ± 436.81	536.63 ± 34.87	Not Provided	Not Provided
Huang 2015 [35]	Gas Injection Group	31	69.94 ± 1.68	3	14	Exclude grade IV and V cataracts	Not Provided	2435.71 ± 194.80	Not Provided	Not Provided	Not Provided
	Fluid Injection Group	27	70.52 ± 1.60	3	15	Exclude grade IV and V cataracts	Not Provided	2430.74 ± 191.95	Not Provided	Not Provided	Not Provided

IABT: intracameral air bubble tamponade; BSS: balanced salt solution; ECD: Corneal endothelial cell density; CCT: Central corneal thickness; CV: Coefficient of variation; HEX: Hexagonal cell percentage

ECD at 1 month postoperatively

Five studies [29–33] that reported the ECD at 1 month postoperatively were included in the analysis. The random-effects model analysis showed that the ECD in the AC air injection group was lower than that in the control group (95% confidence interval [CI]: −256.86 to 49.68; Fig. 2A). The 95% CI for this pooled effect size was 0 (p = 0.19), which indicated no significant between-group difference in postoperative corneal ECD.

Fig. 2.

Forest plots comparing corneal endothelial parameters between the air tamponade and stromal hydration groups at 1 month postoperatively. (A) ECD. (B) CCT. (C) HEX. (D) CV. CCT, central corneal thickness; CV, coefficient of variation; ECD, endothelial cell density; HEX, percentage of hexagonal cells

However, testing revealed high heterogeneity among the included studies (I² = 83%; p = 0.0001). There were contradictions among the results of several studies. The study by Mataftsi et al. [30] showed a significant reduction in ECD in the air-injection group (MD = −506.00), whereas other studies [29, 32, 33] reported no significant between-group differences. Notably, the reduction in ECD in the air-injection group in Mataftsi et al.’s [30] study was observed only in patients operated on by one particular surgeon, which implicated the surgical method as a key confounding factor. These contradictory results were the primary reasons for the high heterogeneity that was observed.

CCT at 1 month postoperatively

In total, four studies [29, 31–33] that reported postoperative CCT findings were included in the analysis. The random-effects model meta-analysis showed that the pooled MD of CCT in the AC of the air-injection group compared to the control group was −4.86 μm (95% CI: −19.81 to 10.08 μm; Fig. 2B). The 95% CI for this pooled effect size was zero (p = 0.52), which indicated no significant between-group difference in the postoperative corneal thickness. Heterogeneity testing revealed high heterogeneity among the included studies (I² = 88%; Tau² = 199.69, p < 0.0001). The results of the individual studies showed prominent inconsistency and even the opposite conclusions: Alsmman et al. [29] showed a significantly thicker CCT in the air-filled group (MD = 12.00 μm), whereas the results of the other three studies [31–33] all indicated a thicker CCT in the control group, with significant between-group differences [31, 32]. These contradictory results were the primary reasons for the high heterogeneity that was observed.

CV at 1 month postoperatively

In total, four studies [29, 31–33] that reported the CV at 1 month postoperatively were included in the analysis. The random-effects model meta-analysis showed that the pooled mean difference in the CV of the AC air-injection group compared to the control group was −0.01 (95% CI: −0.04 to 0.01; Fig. 2C). The 95% CI for this pooled effect size was zero (p = 0.22), indicating no significant between-group difference in the homogeneity of corneal endothelial cell morphology.

Heterogeneity testing revealed substantial heterogeneity among the included studies (I² = 73%; Tau² = 0.00; p = 0.01). This finding suggests considerable variation in the results across different studies. Specifically, Alsmman et al. [29] showed that air filling significantly reduced the CV value (MD = −0.04), whereas the results from Zaki and Mohammad [31], Keles and Karaman [33], and Chen et al. [32] showed no significant between-group differences. The inconsistency between studies contributed to the heterogeneity that was observed.

HEX at 1 month postoperatively

In total, four studies [29, 31–33] that reported HEX levels at 1 month postoperatively were included in the analysis. The fixed-effect model meta-analysis showed that the pooled mean difference in the HEX after AC air-injection group compared to the control group was −0.01 (95% CI: −0.03 to −0.00; Fig. 2D). The 95% CI for this pooled effect size did not include zero (p = 0.03), indicating that the AC Ringer’s solution or BSS filling was statistically superior to AC air filling in promoting the maintenance of regular hexagonal morphology in corneal endothelial cells, and this difference was statistically significant. However, the absolute value of the pooled effect size (−0.01) was small, and its clinical significance should be carefully evaluated in the specific context.

Heterogeneity analysis revealed no heterogeneity among the included studies (I² = 0%, p = 0.44). This indicated that the results for the HEX metric were highly consistent across studies, indicating a trend of slightly poorer cellular regularity in the air-filled group. Among the four studies, the two with the greatest weight [29, 33] showed a trend toward lower HEX values in the air-filled group, which jointly resulted in the final pooled results.

Sensitivity analysis

The robustness of the pooled results was assessed by a leave-one-out sensitivity analysis. For the ECD at 1 month postoperatively, the robustness of the pooled results was generally inadequate. After excluding the study by Mataftsi [30], the heterogeneity among the studies was completely resolved (I² decreased from 83% to 9%), and the pooled results remained statistically non-significant. This indicates that the current conclusion of “no statistical significance” is relatively reliable. The detailed results are shown in Fig. 3A.

Fig. 3.

Leave-one-out sensitivity analysis for the primary outcomes. (A) ECD. (B) CCT. (C) CV. CCT, central corneal thickness; CV, coefficient of variation; ECD, endothelial cell density

For the CCT at 1 month postoperatively, the results indicated that the pooled effect size and heterogeneity were highly unstable. Specifically, after excluding the study by Alsmman et al. [29], the direction of the pooled effect size changed, and heterogeneity decreased substantially (with I² significantly dropping from 87% to 0%), as shown in the results (MD = −10.76, 95% CI: −17.89 to −3.63). These findings suggest that AC air injection potentially exerts a positive effect on reducing corneal edema during the early postoperative period. Thus, the conclusion of “no significant difference” is highly sensitive to this particular study, and indicates poor robustness of the results, which should be interpreted with caution (Fig. 3B).

For CV at 1 month postoperatively, the robustness of the results was generally high (I² = 73%). After excluding the study by Alsmman et al. [29], the heterogeneity was resolved (I² decreased to 38%), and the pooled result remained statistically non-significant. This indicates that the current conclusion of “no statistical significance” is relatively reliable (Fig. 3C).

In summary, the findings of this meta-analysis of the three key outcomes were not robust, thus highlighting the need for more high-quality and methodologically consistent studies to provide clearer evidence.

Risk of bias and quality assessments for the meta-analysis

The ROB-2 tools, which were used to assess the risk of bias in randomized studies, indicated a moderate risk of bias in the included studies, and suggested that the reliability of the randomized studies was moderate. However, the ROBINS-I tool used for non-randomized study evaluation showed a potentially high risk of bias among the included studies, particularly regarding participant selection and confounding factors. Therefore, the results of the present meta-analysis should be interpreted with caution. The risk of bias in the included studies is presented in Figs. 4A and 4B.

Fig. 4.

Assessment of the risk of bias in the randomized and non-randomized controlled trials using the RoB-2 tool (A) and ROBINS-I tool (B) correspondingly. ROBINS-I, risk of bias in non-randomized studies of Interventions; RoB-2, Cochrane risk of bias, Version 2

Summary of adverse events in included studies

Other adverse events and safety outcomes in our included studies such as IOP spikes, anterior chamber reformation, and patient-reported visual symptoms are detailed in Table 2. The air tamponade groups did not show any clinically significant sustained elevation compared to the control groups, nor was there a need to initiate intraocular pressure-lowering treatment owing to air tamponade according to Chen et al. [32] and Huang et al. [35]. Huang et al. [35] reported two cases of patients’ visual symptoms because of bubble migration to vitreous and posterior chamber, but these symptoms were temporary and disappeared as the bubbles were absorbed, without causing long-term effects.

Table 2.

Reporting of adverse events in included studies

Study source	Pupillary Block	IOP Spikes	Need for IOP-Lowering Therapy	Urgent Air Release	Anterior Chamber Reformation Issues	Patient-Reported Visual Symptoms
Keles 2023 [33]	N	N	N	N	N	N
Alsmman 2018 [29]	N	N	N	N	N	N
Zaki 2021 [31]	N	N	N	N	N	N
Chen 2022 [32]	N	During Follow-up period (3 days, 10 days, 1 month, and 3 months), similar to control group	N	N	N	N
Yuksel 2019 [34]	N	N	N	N	N	N
Mataftsi 2020 [30]	N	N	N	N	N	N
Huang 2015 [35]	N	no significant difference on postoperative day 1 compared to preoperative values	N	N	Air group had 94% success in single-operation stabilization (vs. 33% in control)	One case of bubble migration to vitreous causing visual shadow; one case of bubble in posterior chamber; all resolved with positioning

N: Not reported

Discussion

This systematic review and meta-analysis evaluated the effects of AC air tamponade after the completion of phacoemulsification surgery. The primary conclusion of this study was that both the techniques demonstrated comparable short-term corneal endothelial safety. The meta-analysis revealed no significant between-group differences regarding ECD, CCT, and CV 1 month postoperatively. While the pooled analysis showed a statistically significant 1% reduction in hexagonality favoring the no-air group (−0.01 [−0.03 to −0.00]; p = 0.03), the clinical significance of this minimal difference remains questionable.

Notably, leave-one-out analysis demonstrated that the exclusion of Mataftsi et al. [30] completely resolved heterogeneity for endothelial cell density (I² decreased from 83% to 9%), suggesting that the ‘no significant difference’ conclusion is highly dependent on study inclusion criteria and individual study characteristics. Significant heterogeneity was observed in the CCT and CV outcomes, primarily stemming from the divergent results reported by Alsmman et al. [29]. The discrepancies in this study might be attributed to the relatively lower phacoemulsification energy, and thus reduced the overall surgical trauma compared to other studies. Sensitivity analysis confirmed that the pooled results for CCT were unstable, underscoring existing heterogeneity. This heterogeneity may have originated from variations in air injection protocols, differences in surgical techniques, diverse patient baseline characteristics, and potential inconsistencies in the corneal endothelial specular microscopy measurement methods across the included studies.

The included studies primarily focused on corneal biometric parameters and lacked a comprehensive assessment of complications such as transient IOP elevation or pupillary block. The incomplete reporting of clinically important safety outcomes represents a significant evidence gap and limited our meta-analysis. Pupillary block is a recognized risk of anterior chamber air, as documented in cataract surgery [36]. Flowers et al. described a case of pupillary block and angle-closure glaucoma following anterior chamber air injection in a nanophthalmic eye [36]. In a commentary, Shugar reported two similar cases in non-nanophthalmic eyes among 1123 consecutive cataract surgeries, noting that excessive bubble size was the common denominator and that prompt decompression by “burping” the incision effectively relieved the attack [36]. Future studies are needed to incorporate prespecified safety endpoints including pupillary block and IOP curves postoperatively.

According to some researches, AC injections seem to offer additional benefits other than endothelial safety (Table 3). Keles and Karaman [33] found a lower CCT in the air-injection group on the first postoperative day (indicating less edema). Thus, this may help prevent DMD, as Ti et al. [42] confirmed that preexisting endothelial pathology and corneal edema on the first postoperative day are independent risk factors for DMD. Chen et al. [32] indicated that this technique significantly shortened surgical time and reduced the incidence of persistent postoperative corneal edema. Demirci et al. [38] found that the bubble lowers aqueous nitric oxide levels and reduces AC inflammation in rabbit eyes. Mehdizadeh et al. [14] further showed that the bubble significantly decreases the incidence of Staphylococcus epidermidis-induced endophthalmitis. However, these potential benefits of reduced postoperative inflammation and endophthalmitis come from animal experiments and lack clear clinical trials to verify them, and thus, this is only a speculation. In the future, large-scale clinical trials are needed to validate this speculation. Yuksel et al. [34] confirmed that air injection markedly lessens early postoperative complications such as DMD, endothelial wound gape, and epithelial wound gape.

Table 3.

Potential effects other than corneal endothelial safety reported in literature

First Author & Year	Study Type	Subjects	Research Focus/Direction	Total Sample Size	Main Conclusion
Sim 2007 [37]	Case Series	Human	To evaluate if an anterior chamber air bubble can prevent inflow of ocular surface fluid through sutureless cataract wounds.	14 eyes	Injecting a small air bubble at the end of surgery effectively prevented the inflow of ocular surface fluid through the clear corneal incision, both spontaneously and upon manipulation.
Demirci 2013 [38]	Experimental study	Rabbit	To investigate the effect of an intracameral air bubble on postoperative inflammation using clinical grading and nitric oxide levels.	30 eyes (15 rabbits)	Air bubble significantly reduces post-cataract inflammation.
Mehdizadeh 2009 [14]	Experimental study	Rabbit	To assess the ability of an anterior chamber air bubble to prevent the development of experimental bacterial endophthalmitis.	24 eyes (24 rabbits)	Air bubble reduces the incidence of S. epidermidis endophthalmitis.
Werner 2015 [39]	Laboratory case series	Human	To describe and analyze the phenomenon of localized calcification on hydrophilic acrylic IOLs following surgeries involving intracameral air/gas injection (e.g., DSEK/DSAEK).	7 cases (7 IOLs)	Intracameral air/gas can cause localized calcification on the anterior surface of hydrophilic IOLs, impairing vision.
Zhou 2025 [40]	Retrospective Cohort Study	Human	To investigate the impact of anterior chamber air injection (ACAI) on IOL stability (decentration, tilt, position) after combined cataract and vitrectomy surgery with gas/air tamponade.	51 eyes (51 patients)	ACAI did not cause significant IOL decentration or tilt but led to a significant anterior axial shift of the IOL and prolonged the stabilization time of aqueous depth.
Leisser 2019 [41]	Randomized controlled monocenter study	Human	To examine whether air tamponade induces more postoperative IOL tilt or decentration compared to balanced salt solution in phacovitrectomy.	34 eyes (from 42 initial)	Air tamponade does not induce significantly more IOL tilt or decentration compared to balanced salt solution.

Concerns have been raised regarding the potential adverse effects of AC air injections. Although Zhou et al. [40] indicated that AC air injection during phacovitrectomy does not exacerbate IOL tilt or decentration, it has been established that air injection affects the axial position of the IOL and causes an anterior shift. This phenomenon is particularly evident in specific patient groups, such as males and those with a greater preoperative aqueous width. Moreover, Werner et al. [39] indicated a risk of localized calcification and reduced visual acuity associated with hydrophilic acrylic IOLs following AC air injection, suggesting that such IOLs should be avoided when employing this technique (Table 3).

While our meta-analysis suggests comparable endothelial safety, we caution that this interpretation is constrained by significant heterogeneity (I² > 70% for multiple outcomes) and the predominance of short-term follow-up data. The absence of statistically significant differences should not be interpreted as proof of equivalence but should rather be interpreted as uncertainty requiring further investigation. Furthermore, this meta-analysis has a few limitations, including the limited number of included studies, primary focus on short-term follow-up at 1 month postoperatively, absence of hard nuclear cataracts in the included studies, and primary focus on corneal biometric parameters without a comprehensive assessment of other potential complications such as transient IOP elevation or visual disturbances. Future research should prioritize large-scale, methodologically uniform, and RCTs with longer follow-up durations. These studies must standardize the air injection protocol and broaden the outcome measures to include patient-reported visual quality, intraocular lens position, postoperative inflammatory response, IOP fluctuations, and incidence of rare but serious complications such as endophthalmitis. This approach could enable a more comprehensive assessment of the risks and benefits of routine AC air injection during cataract surgery.

Conclusions

This study demonstrates that, at 1-month postoperatively, intracameral air injection exhibits comparable efficacy to BSS in terms of ECD, CCT, and CV. Although a statistically significant reduction in HEX was observed in the air group (approximately 1%), its clinical relevance remains limited. It should be noted that substantial heterogeneity was present in some outcomes, and sensitivity analyses indicated instability in the pooled estimates for parameters such as CCT, warranting cautious interpretation. The included literature lacks systematic reporting of critical safety outcomes, particularly pupillary block and IOP-related events, which are essential for assessing the net clinical benefit of this technique. While theoretical advantages of air tamponade may exist, they have not been robustly demonstrated in clinical comparative studies. Well-designed randomized controlled trials with standardized protocols, longer follow-up, and comprehensive safety assessment are urgently needed to guide clinical practice.

Abbreviations

BSS

Balanced salt solution

CCT

Central corneal thickness

RoB-2

Cochrane Risk of Bias, Version 2

CV

Coefficient of variation

DMD

Descemet’s membrane detachment

ECD

Endothelial cell density

AC

Anterior chamber

HEX

Hexagonality

IOL

Intraocular lens

IOP

intraocular pressure

MD

Mean difference

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

RCTs

Randomized controlled trials

ROBINS-I

Risk of Bias in Non-randomized Studies of Interventions

Authors’ contributions

XBG: Conceptualization, Data curation, Formal analysis, Writing – original draft. QL: Data curation, Formal analysis, Methodology. FRL: Investigation, Validation. LJ: Investigation, Resources. LY: Supervision, Writing – review & editing. JYZ: Conceptualization, Project administration, Supervision, Writing – review & editing, Funding acquisition.

Funding

This study was supported by the National Natural Science Foundation of China [82201191] and Natural Science Foundation of Hubei Province [2023AFB461].

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.