Blood Perfusion in Rotational Full-Thickness Lower Eyelid Flaps Measured by Laser Speckle Contrast Imaging

Purpose:

Large upper eyelid defects can be repaired by rotational full-thickness lower eyelid flaps. The aim was to measure the blood perfusion in such flaps, and how it is affected by the length of the flaps, and the degree of rotation and stretching.

Methods:

Nine patients underwent the Quickert procedure for entropion repair in which a full-thickness eyelid flap of approximate width 0.5 cm and length 2 cm was dissected in the lower eyelid. This generates a full-thickness eyelid flap similar to that used to repair large upper eyelid defects. Perfusion was measured using laser speckle contrast imaging, before and after the flap was rotated 90° and 120°, and stretched using forces of 0.5, 1, and 2 N.

Results:

Blood perfusion decreased gradually from the base to the tip of the flap; being 75% of the reference value 0.5 cm from the base, 63% at 1.0 cm, 55% at 1.5, 23% at 1.75 cm, and 4% at 2.0 cm. Rotating the flaps by 90° or 120° had little effect on the perfusion. Stretching reduced the perfusion from 63% to 32% at 2 N, when measured at 1 cm. The combination of stretching and rotation did not lead to any further decrease.

Conclusions:

Blood perfusion in lower eyelid rotational flaps seems to be more sensitive to stretching than to rotation. Provided the flap is no longer than 1.5 cm, the perfusion will be greater than 20%, even when rotated, which should be sufficient for adequate survival and healing.

Blood perfusion in lower eyelid rotational flaps in patients will be greater than 20%, provided the flap is no longer than 1.5 cm, even when rotated.

A number of surgical procedures can be used to repair full-thickness eyelid defects involving the eyelid margin, including direct closure, flaps, tissue grafts, or a combination of these.¹ Direct closure can be used to repair full-thickness eyelid defects of less than 25% of the width of the eyelid. However, when the removal of a tumor affects more than 50% of the width of the eyelid, it is necessary to repair the eyelid using a flap or graft. A local flap is preferred because this has similar texture, thickness, and color, it may also provide eyelashes for the defect, and there is less contraction on healing than with skin grafts.² One technique that may be used for this is a full-thickness eyelid flap. For example, a large upper eyelid defect can be repaired by rotating the lower eyelid into the upper eyelid defect.³

It is well known that the survival of a flap is affected by its design. An adequate blood supply to the tissue is crucial for the healing of both skin flaps and grafts. However, the design of flaps is largely based on empirical observations of flap survival and information available in the literature.⁴ Methods used previously to measure blood perfusion during surgery have not been reliable. Fluorescein and disulfine blue dye have been used to measure perfusion and flap survival, but these methods are associated with underestimation or overestimation of flap survival by up to 30%.⁵ Recently developed techniques now allow easier and more reliable measurements of blood flow in plastic surgery. Laser speckle contrast imaging (LSCI) is a noninvasive technique that provides high-resolution images of the structure and function of the tissue with high reproducibility.⁶ The object is illuminated by laser light, and the backscattered light, which forms a random interference pattern called a speckle pattern, is used to determine perfusion.⁷ Movement, such as the flow of red blood cells in a tissue, causes the speckle pattern to change, allowing the blood perfusion to be quantified. Laser speckle contrast imaging is a fast, full-field technique for the imaging of microvascular perfusion.⁸ Current LSCI equipment can produce representative images of the perfusion in the surface of tissue over a relatively large area (up to 24 × 24 cm). Laser speckle contrast imaging is now an established technique in, for example, experimental research and plastic surgery.^9–13

The aim of this study was to measure blood perfusion using LSCI in full-thickness rotational eyelid flaps in humans and to investigate how perfusion is affected by the length of the flap, the degree of rotation, and stretching of the flap. This is the first time that perfusion has been measured in a full-thickness rotational eyelid flap during surgery in humans.

METHODS

Ethics.

The experimental protocol for this study was approved by the Ethics Committee at Lund University, Sweden. The research adhered to the tenets of the Declaration of Helsinki as amended in 2008. All patients included in the study gave their fully informed consent.

Subjects.

Patients admitted to the Department of Ophthalmology, Skåne University Hospital, for surgery to repair involutional entropion were consecutively recruited for the study during the period January to May 2018. All patients had excessive eyelid laxity and were scheduled for entropion surgery using the Quickert procedure.¹⁴ The patient’s medical history was unknown before the day of surgery. Exclusion criteria were inability to provide informed consent, and physical or mental inability to cooperate during the local anesthetic procedure. No patients were excluded. Nine eyelids in 8 patients were included in the study. The patient’s characteristics are given in the Table.

Patient’s characteristics

Surgical Procedure.

Surgery was performed under local infiltration anesthesia with 20 mg/ml lidocaine (Xylocaine, AstraZeneca, Södertälje, Sweden). A modified Quickert procedure was performed,¹⁴ in which a vertical incision was made through the eyelid, and through the tarsal plate, approximately 5 mm from the lateral canthus. A horizontal full-thickness incision was then made below the lower border of the tarsal plate. In this way, 2 full-thickness eyelid flaps were created, of which the medial one was the longer, and in which perfusion was measured. The median dimensions of the flaps were 0.5 cm (0.4–0.6) vertically and 2.1 cm (1.8–2.5) horizontally, and the median thickness was 0.3 cm (0.2–0.4). Perfusion measurements were performed 10 minutes after the creation of the flap to minimize the effect of surgical vasospasm. After perfusion measurements had been made, the distal part of the medial flap was resected to correct the excess eyelid laxity as part of the modified Quickert procedure. The Quickert procedure was then completed to correct the entropion. There were no adverse events, and the surgical results were good in all cases.

Laser Speckle Contrast Imaging.

Blood perfusion was measured using LSCI (PeriCam PSI NR System, Perimed AB, Stockholm, Sweden). This system employs an infrared 785 nm laser beam that is spread over the surface of the skin by a diffuser, creating a speckle pattern (dark and bright areas formed by random interference of the light backscattered from the illuminated area). Blood perfusion is calculated automatically by the system by analyzing the variations in the speckle pattern. The speckle pattern is recorded in real time, at a rate of up to 100 images per second, with a high resolution of up to 100 μm/pixel. The technique has been used to monitor flap perfusion in other plastic reconstructive surgery procedure before both in humans and porcine models.^9–13 It is well known that the skin perfusion is affected by cardiovascular status and the room temperature. The room temperature was standardized to 20°C. To conquer the problem with variability in perfusion between the subjects, the results were calculated as percent change (see Calculations and Statistics).

Study Protocol.

A medium-sized corneal shield was used to protect the eye from laser irradiation (Ellman International Inc., Oceanside, NY). A thin plastic shield was applied under the flap to prevent interference due to the laser signal resulting from blood flow in the underlying tissues. Perfusion was imaged over the entire length of the flap. Perfusion was measured at the base of the flap (0 mm) before rotating or stretching the flap and was set to 100%. Perfusion was measured in the resected part of the flap and set to 0%. Perfusion was then measured every 0.25 cm along the length of the flap, without any rotation or stretching of the flap. The flaps were then rotated manually by 90° and 120° and the perfusion determined again. Two sutures were attached to the distal end of the flap to stretch the flaps. The sutures were then attached to a line, which was threaded through a fixed pulley. Different weights were added (50, 100, and 200 g) to achieve stretching forces of 0.5, 1, and 2 N. The stretched flaps were then also rotated. One limitation of the present study was that the perfusion was not measured at different time points after completion of a rotational flap. The reason is that these were modified Quickert procedures, the flap was shortened at the end of surgery and not sutured in place rotated. The relevance of postsurgical measurements would therefore be limited. Furthermore, performing these measurements at different time points after the creation of the flap would be unethical because it would extend the duration of the surgery. However, in a future study on true rotational flaps, this could be done after surgery has been completed.

For the completion of the entropion repair, part of the medial flap is resected to redress excess eyelid laxity. Perfusion was measured in the resected part of the flap to obtain a value of zero perfusion. After completion of the study protocol, the surgical procedure to correct entropion was performed according to normal practice.

The design of the flap and the perfusion measurements are illustrated in Figure 1.

FIG. 1.

Illustration of the full-thickness eyelid flap. Perfusion was measured using LSCI at 0.25 cm increments from the base of the flap. Comparisons between measurements with and without rotation and tension were made at a distance of 1 cm. LSCI, laser speckle contrast imaging.

Calculations and Statistics.

Perfusion measured with LSCI is expressed in arbitrary units, perfusion units. Blood perfusion was calculated and expressed as a percent (median values and interquartile ranges) after normalization to the perfusion at the base of the flap (100%) and in the resected part (0%). Statistical analysis was performed using the Kruskal–Wallis test and Dunn’s multiple comparison test. Significance was defined as p < 0.05 and p > 0.05 (not significant). All differences referred to in the text were statistically verified. Calculations and statistical analysis were performed using GraphPad Prism 7.0a (GraphPad Software Inc., San Diego, CA).

RESULTS

Blood perfusion decreased gradually from the base to the tip of the flap; being 75% of the reference value 0.5 cm from the base, 63% at 1.0 cm, 55% at 1.5, 23% at 1.75 cm, and 4% at 2.0 cm. Beyond 2.0 cm there was only minimal perfusion (Fig. 2A).

FIG. 2.

Blood perfusion measured using LSCI in full-thickness rotational eyelid flaps, showing the effects of distance from the base (0–2 cm) (A), rotation (90° and 120°) (B), and stretching of the flap (0.5, 1, and 2 N) (C). Data are shown as medians and interquartile ranges. Statistical analysis was performed using Friedman’s test with Dunnett’s multiple comparison test. Significance was defined as p < 0.05 and p > 0.05 (n.s.). Note that rotating the flaps had little effect on the perfusion while stretching the flaps significantly reduced the perfusion. LSCI, laser speckle contrast imaging; n.s., not significant.

Rotating the flaps by 90° or 120° had little effect on the perfusion (Figs. 2B and 3). Stretching the flaps with a force of 0.5 or 1 N, with no rotation, resulted in a nonsignificant decrease in perfusion (from 63% to 56%, and 40%, 1.0 cm from the flap base, p = not significant). When the higher force of 2 N was applied, the perfusion decreased to 32%, p < 0.01 (Figs. 2C and 4). At this force, the perfusion was lower along the entire length of the flap than in the unstretched flap.

FIG. 3.

Schematic illustration of flap rotation (left), together with photographs (top row), and corresponding laser speckle images (bottom row) showing the effects of rotation of a full-thickness eyelid flap by 90° and 120°. Purple color, Low blood perfusion; yellow/white color, high blood perfusion. It can be seen that rotating the flaps had little effect on the perfusion. PU, perfusion units.

FIG. 4.

Schematic illustration of flap stretching (left), together with LSCI images showing perfusion in a full-thickness eyelid flap stretched by 0.5, 1, and 2 N. Purple color, Low blood perfusion; yellow/white color high blood perfusion. It can be seen that stretching the flaps significantly reduced the perfusion. LSCI, laser speckle contrast imaging; PU, perfusion units.

No significant differences in perfusion were seen when stretching an already rotated flap. The perfusion when applying a force of 1 N to the unrotated flap was 42%, compared with 51% and 49% when the flap was rotated by 90° and 120°, respectively (measured 1 cm from the flap base). At the highest force of 2 N, the perfusion decreased to 29%, 30% and 28% in the unrotated, and in the 90° and 120° rotated flaps, respectively (Fig. 2).

DISCUSSION

The results of the present study show a decrease in perfusion along the length of the full-thickness rotational eyelid flaps in patients. Perfusion was found up to a distance of 1.5 cm from the base (~30%), but beyond this perfusion was minimal. This indicates that 1.5 cm is probably the maximum length for a rotational flap, with retained perfusion, especially as it was found that stretching reduced perfusion even further. However, in clinical practice, such flaps often must be longer to repair large defects. In cases where the tip of the flap survives in longer flaps, this is presumably due to the passive diffusion of oxygen and nutrients. The authors have indeed recently shown that a free full-thickness eyelid wedge graft 1 cm wide can be transposed to another eyelid without necrosis.¹⁵

Previous studies support the findings of perfusion being dependent on the length of the flap.^9–11 The perfusion in full-thickness eyelid flaps in pigs¹¹ has been found to be higher than in this study in patients; being 80% 3 cm from the base of the flap. This may be because the pig eyelid is thicker, having a more extensive vascular network. Furthermore, some of the patients in the present study suffered from hypertension and/or cardiovascular disease, which may compromise blood flow. The mean age of the patients was 83 years, which may also have affected blood flow. Moreover, these patients were undergoing surgery for entropion, and the tissue may have been affected by inflammation or edema, both of which could compromise wound blood flow and/or healing.

Interestingly, perfusion in these full-thickness eyelid flaps was found to be better than in random pattern flaps.¹⁵ This may be because the eyelid can be regarded as an axial flap, or arterial flap, which is a myocutaneous flap containing a direct cutaneous artery along its longitudinal axis. A random skin flap, however, lacks a specific vessel for vascularization and is perfused by musculocutaneous microcirculation in the tissue.¹⁶

Advancing a flap to cover a defect following tumor excision or trauma often requires stretching and/or rotation of the flap. When the flaps were stretched with a force of 2 N, perfusion decreased by 50%. The authors believe that a force of 2 N may be at the upper limit of that used clinically. However, the level of stretch that is seen clinically cannot be deduced from the present study, and there are no studies to confirm that 2 N may be clinically relevant. The true effect of stretch on a flap thus remains unknown. Rotation of the flap did not have any significant effect on perfusion in the present study. It is not unnatural to assume that rotating a skin flap would impair blood perfusion by strangulation of the blood vessels. In a previous study, in which perfusion was measured in random skin flaps in a porcine model, the perfusion was found to decrease significantly when the flap was manipulated by stretching and/or rotating it 90°.¹⁰ This may be due to differences in the structure of porcine and human skin.

One limitation of the present study is that blood perfusion was measured in a flap created for entropion repair. It was therefore not sutured into place, as would a full-thickness flap for the repair of a defect after tumor surgery, making long-term follow-up impossible. It is therefore not possible to conclude what degree of perfusion would be adequate for survival or healing of the flap. Neither could flap necrosis be evaluated. However, it is well known that eyelid flaps have a high survival rate, probably due to rich periorbital vascularization that benefits the diffusion of both oxygen and nutrients from the adjoining tissue and from the tear fluid. According to Tyers and Collin,¹⁷ many flaps survive even when they appear discolored during the first few days postoperatively. Another limitation of this study is that the base of all the flaps was medial. Indeed, there is generally a low risk of flap necrosis in eyelid surgery, so the usefulness of measuring perfusion for this particular procedure would be unlikely to change clinical management. It is well known that most of the blood flow to the eyelid is from the medial canthus,¹ and it can therefore not be ruled out that the results may have been different if the flap had been dissected extending from the lateral canthus, or in the upper eyelid. Another limitation is the sample size. Nine patients are just barely enough to allow statistical analysis of significance. However, the results display a clear-cut trend in decreased perfusion along the length of the flap and the results indeed reach statistical significance.

In conclusion, full-thickness rotational eyelid flaps are perfused in the proximal 1.5 cm, and the survival of longer flaps will be dependent on the passive diffusion of oxygen and nutrients. The application of tension to a flap reduced the perfusion, while rotation up to 120° did not seem to affect perfusion. Considering all the findings of this study, we conclude that a flap must be sufficiently long to allow it to be moved from the donor site to the recipient site, without applying too high a force, and it should be short enough to ensure adequate perfusion of the tip.

ACKNOWLEDGMENTS

We thank Helen Sheppard for her valuable help with the language, and all the surgical staff involved at Skåne University Hospital.