The Yin and Yang of Primary Unilateral Cleft Lip and Nose Repair: Balance through Opposing Cleft and Noncleft Side Changes

Abstract

Background: 

Balance is achieved through opposing interactions. Objective three-dimensional assessment of changes during surgical treatment of the unilateral cleft lip and nasal deformity are limited, and false assumptions may prevent optimal management.

Methods: 

The authors performed anthropometric analysis on the immediate preoperative and postoperative images (captured under anesthetic) of patients undergoing primary repair (n = 36). Changes in dimensions and measures of balance were assessed (P < 0.05).

Results: 

Angles and ratios that reflect cleft to noncleft side balance normalized, although alterations occurred in opposing ways. Centralization of the columella narrowed the cleft nasal base and widened the noncleft nasal base. As the cleft columellar height elongated, the noncleft columellar height shortened. With these changes and correction of cleft alar base retrusion, the cleft alar dome was raised. The cleft and noncleft lateral lip heights and widths elongated. Meanwhile, the Cupid’s bow broadened as the commissures were drawn closer together. Whereas the cleft philtral height lengthened, the noncleft philtral height shortened. Reduction in noncleft philtral height averaged 20% but varied with measures of preoperative severity including columellar angle (R = 0.67), the difference in philtral heights (R = 0.65), and lateral deviation of the subnasale (R = 0.74).

Conclusions: 

Tissue does not need to be added to “lengthen” the columella, the noncleft philtral height shortening can be estimated, and the contours of anatomic subunits change with surgery on both cleft and noncleft sides. It is inadequate to focus on correction of the cleft side alone without considering corresponding noncleft side changes. Achieving balance through opposing alterations should be the principal goal of treatment.

CLINICAL QUESTION/LEVEL OF EVIDENCE: 

Therapeutic, IV.

Fundamental to any reconstruction is a clear understanding of the deformity¹ and the changes that occur with treatment. The unilateral cleft lip nasal deformity (uCLND) is complex, and “optical illusions” can lead surgeons to false assumptions and incorrect goals.^2,3 Although classic descriptions of uCLND emphasize lateral deviation of the cleft side alar base^4–16 and consider the noncleft side to be the “ideal normal,”^17,18 objective analysis has revealed otherwise.¹⁹ Uncoupled growth of the premaxilla²⁰ results in an outward swing of the anterior nasal spine^19,21 with consequent deviation of columella and noncleft alar base away from midline.^19,21,22 Meanwhile, deficiency of the lesser segment²¹ results in retrusion of the cleft alar base,^19,21,22 such that the orthogonal displacements result in a twisting collapse of the nose.¹⁹ Using objective analysis of three-dimensional images, we have found that these changes occur in a dependent manner to the lateral deviation of the anterior nasal spine¹⁹ and that the cleft alar base is normal in mediolateral position.²²

We also found that opposing cleft and noncleft side deformities occur, and worsen, through the spectrum of presentation: as the cleft medial lip height diminishes, the noncleft medial lip height elongates; as the cleft nasal sill widens, the noncleft nasal sill narrows; and as the cleft alar dome collapses, the noncleft alar dome rises to become more “boxy.”^19,23 Although noncleft side deformities may be apparent to some surgeons, they have not been well described or documented. These anomalies are less obvious because our attention tends to be drawn to the cleft side, where form in this study appears less familiar. Nonetheless, if the unrepaired uCLND involves an interplay of opposing changes, it follows that correction should include cleft and noncleft side alterations. Assessment of these relationships is critical to practice because contemporary techniques focus solely on cleft side adjustments, rely on noncleft side dimensions for surgical planning, and assume that the noncleft side remains static through surgery. The purpose of this study was to assess three-dimensional alterations that are produced with primary cleft lip repair and to understand how noncleft side changes contribute to nasolabial balance.

Although our previous study demonstrated that balance could be attained and maintained through 5 years of age,²⁰ given the longitudinal nature and necessity of in-clinic image capture, confounders from growth and facial animation precluded determination of relative cleft and noncleft side changes. We therefore focused on immediate intraoperative form so that we could determine “how” balance was achieved. By isolating direct surgical changes, we could “deconstruct” the repair and better understand mechanisms that contribute to balance.

PATIENTS AND METHODS

We reviewed consecutive patients undergoing primary unilateral cleft lip repair over a 2-year period (n = 49). Thirty-six had available three-dimensional intraoperative images that were acquired while patients were asleep under general anesthetic. We used the Vectra H1 imaging system (Canfield Scientific, Fairfield, NJ). Surgical repairs were performed by a single surgeon using the anatomic subunit approximation technique for cleft lip repair^24,25 with foundation-based primary rhinoplasty (septoplasty, nasal floor closure, and lateral nasal wall reconstruction, as needed), with no nasal tip dissection.²⁶ Apart from muscle release from the columellar base and periosteum release along the noncleft nasal floor (to centralize the nasal septum),²⁶ these approaches do not involve any dissection or direct manipulation of the noncleft side.

Images were normalized to x, y, and z axes, as described previously,¹⁹ with the facial midline midway between endocanthion points. Landmarks were placed on preoperative and postoperative images of each subject using 3D Slicer (version 4.11.20210226).²⁷

Anthropometric Analysis

Analysis was as described previously (Fig. 1).^28,29 Columellar angle was the planar angle off of a perpendicular to the endocanthion-endocanthion horizontal. Columellar width was measured at the narrowest region of the columella. Columellar height, on each side, was measured from the nasolabial crease to the inflection between the vertical and horizontal contours along the nostril rim.^30–32 Nostril width was the maximum horizontal dimension within the nostril. Nasal base width differs and was defined as the absolute distance between subnasale and subalare. Although medial lip height (subnasale to crista philtra inferioris) provides an indirect measure of philtral morphology, we were specifically interested in the height of the philtral columns. We therefore directly measured philtral height, defined as the distance between the crista philtri superioris and the crista philtri inferioris, above the cutaneous roll. Indices of nasolabial balance were expressed as ratios of cleft to noncleft side dimensions, with values closer to 1 indicating greater symmetry.

Fig. 1.: Anthropometric measurements and immediate intraoperative changes with primary repair. (Left) Dimensions measured: nasal width (NW), lip width (LW), alar height (AH), nasal base width (BW), philtral height (PH), lateral lip height (LLH), lateral lip width (LLW), Cupid’s bow width (CBW), columellar angle (CA), columellar height (CH), columellar width (CW), nostril width (NsW), and nostril height (NsH). (Right) Change in millimeters (degrees for columellar angle) from preoperatively to postoperatively (P < 0.05). Note that the measured philtral height between the crista philtri inferioris and the crista philtri superioris does not include the cutaneous roll, as we consider the cutaneous roll a distinct subunit.

Qualitative Audit

To verify the findings of anthropometric analysis and to provide potential explanations for their occurrence, two- and three-dimensional images of all subjects were systematically audited.

Statistical Analysis

We performed morphometric analysis of linear measurements, angles, and ratios in R.³³ Normality of the measurements was assessed using the Kolmogorov-Smirnov test. We compared preoperative and postoperative measures using a paired t test. We assessed the correlation of philtral height changes with preoperative measures of severity using Pearson correlation. Changes associated with a value of P < 0.05 were considered significant.

RESULTS

Patient demographics are summarized in Table 1. The mean age at surgery was 8.1 ± 3.2 months. Surgery resulted in centralization of subnasale, medial advancement of the noncleft alar base, and anterior advancement of the cleft alar base. Anthropometric analysis on 72 images is summarized in Table 2 and changes are summarized in Figure 1.

Table 1. - Patient Demographics

	No. (%)
Sex
Female	10 (27.8)
Male	26 (72.2)
Race
Asian	5 (13.9)
African American	3 (8.3)
White	22 (61.1)
American Indian and Alaska Native	1 (2.8)
Hispanic	4 (11.1)
Other	1 (2.8)
Side of cleft
Left	19 (52.8)
Right	17 (47.2)
Type of clefta
Microform	1 (2.8)
Incomplete	15 (41.7)
Complete cleft with band	3 (8.3)
Complete	17 (47.2)
Extent of complete clefting
Lip	7 (19.4)
Lip and alveolus	14 (38.9)
Lip, alveolus, and palate	15 (41.7)
Presurgical orthopedics	10 (58.8)b
Nasoalveolar molding	4 (23.5)b
Lip taping	6 (35.3)b
Associated syndromes and other anomalies	5 (13.9)
Van der Woude	1 (2.8)
1p36 deletion with nasal dysplasia	1 (2.8)
Craniofacial microsomia	1 (2.8)
Bilateral microtia with midline deficiency	1 (2.8)
Camptodactyly	1 (2.8)

^aComplete cleft with band is defined as a cleft that has a bridge of skin and subcutaneous tissue without any muscle, which is differentiated from an incomplete cleft by a complete cleft of the alveolus (Da Silva Filho OG, Santamaria M, Da Silva Dalben G, Semb G. Prevalence of a Simonart’s band in patients with complete cleft lip and alveolus and complete cleft lip and palate. Cleft Palate Craniofac J. 2006;43:442–445). Microform cleft lip is a mild form of an incomplete cleft lip, and is defined as a cleft that is less than one-fourth of the total height of the lip (Onizuka T, Ichinose M, Hosaka Y, Usui Y, Jinnai T. The contour lines of the upper lip and a revised method of cleft lip repair. Ann Plast Surg. 1991;27:238–252).

^bPercentage of patients with complete clefts who underwent the described presurgical orthopedic treatment.

Table 2. - Intraoperative Three-Dimensional Anthropometric Analysis^a

Measurement	Preoperative	Postoperative	Change	P
Balance
Nose
sn_x, mmb	5.11 (0.42)	0.28 (0.24)	−4.83 (2.16)	<0.0001d
Columellar angle, degc	30.00 (2.21)	1.69 (0.24)	−28.32 (12.98)	<0.0001d
Columellar height ratio	0.52 (0.03)	1.06 (0.03)	0.54 (0.28)	<0.0001d
Nostril width ratio	1.59 (0.08)	0.95 (0.03)	−0.64 (0.41)	<0.0001d
Nasal base width ratio	1.93 (0.08)	0.97 (0.02)	−0.97 (0.49)	<0.0001d
Alar height ratio	0.84 (0.02)	1.02 (0.02)	0.17 (0.13)	<0.0001d
Lip nose ratio	1.17 (0.02)	1.20 (0.02)	0.04 (0.04)	<0.0001d
Lip
Philtral height ratio	0.48 (0.01)	0.98 (0.01)	0.50 (0.12)	<0.0001d
Medial lip height ratio	0.61 (0.02)	1.01 (0.02)	0.40 (0.15)	<0.0001d
Lateral lip height ratio	0.76 (0.03)	0.87 (0.02)	0.11 (0.20)	0.0026d
Lateral lip width ratio	0.86 (0.02)	0.92 (0.01)	0.06 (0.08)	0.0001d
Specific dimensions, mm
Nose
Noncleft columellar height	3.51 (0.13)	3.00 (0.10)	−0.50 (0.77)	0.0004d
Cleft columellar height	1.81 (0.12)	3.16 (0.11)	1.34 (0.86)	<0.0001d
Columellar width	4.08 (0.12)	3.91 (0.11)	−0.16 (0.45)	0.0375d
Noncleft nostril width	5.78 (0.18)	6.08 (0.15)	0.30 (0.64)	0.0076d
Cleft nostril width	8.96 (0.38)	5.74 (0.16)	−3.22 (2.31)	<0.0001d
Noncleft nasal base width	9.02 (0.25)	10.32 (0.22)	1.30 (1.05)	<0.0001d
Cleft nasal base width	16.99 (0.56)	9.89 (0.19)	−7.10 (3.20)	<0.0001d
Noncleft alar height	11.67 (0.19)	11.80 (0.19)	0.13 (0.64)	0.2206
Cleft alar height	9.80 (0.24)	11.93 (0.18)	2.13 (1.35)	<0.0001d
Nasal width	31.01 (0.51)	28.63 (0.37)	−2.38 (1.92)	<0.0001d
Lip
Noncleft philtral height	10.62 (0.27)	8.60 (0.23)	−2.03 (1.06)	<0.0001d
Cleft philtral height	5.09 (0.22)	8.37 (0.21)	3.28 (1.03)	<0.0001d
Noncleft medial lip height	8.69 (0.27)	9.46 (0.26)	0.77 (1.19)	0.0004d
Cleft medial lip height	5.23 (0.18)	9.45 (0.22)	4.23 (1.11)	<0.0001d
Noncleft lateral lip height	10.43 (0.23)	10.91 (0.23)	0.48 (0.93)	0.0043d
Cleft lateral lip height	7.82 (0.25)	9.42 (0.22)	1.59 (1.74)	<0.0001d
Noncleft lateral lip width	19.80 (0.49)	22.09 (0.39)	2.29 (1.63)	<0.0001d
Cleft lateral lip width	16.90 (0.36)	20.32 (0.36)	3.41 (1.40)	<0.0001d

^aAll linear measurements are expressed in millimeters (SD) and ratios are expressed as cleft-to-noncleft.

^bsn_x is the distance of subnasale from endocanthion–endocanthion midline along the x axis (in millimeters).

^cColumellar angle is the deviation of columellar axis viewed from submental view from perpendicular to endocanthion–endocanthion (in degrees).

^dStatistically significant.

Postoperative indices of balance, including lateral deviation of subnasale, columellar angle, and ratios of cleft to noncleft values (nostril width, alar height, and columellar height) normalized. As emphasized during surgical repair, postoperative philtral height ratio normalized to 0.98, whereas the lateral lip width and lateral lip height ratios improved but were further from normal (0.92 and 0.87). Although normalization of ratios was expected with surgery, we found that the changes were not limited to the cleft side, but rather opposing cleft and noncleft side changes contributed to the improvements in nasolabial balance (Fig. 1).

Opposing Nasal Changes (P < 0.05)

The cleft nostril and nasal base widths narrowed; however, this was not because of medial advancement of the cleft alar base. Rather, the narrowing occurred with centralization of columella to facial midline (Fig. 2). Although the noncleft alar base moves medially with the columella,²² the shift in noncleft alar base was not as great. Consequently, the noncleft nostril and nasal base widths widened (+0.3 and +1.3 mm), while the corresponding cleft side dimensions narrowed (−3.2 and −7.1 mm).

Fig. 2.: Opposing changes. Uncoupled premaxillary growth results in outward swing of the anterior nasal spine and lateral deviation of the columella and noncleft side alar base from midline (blue vertical). During repair, as the columella is centralized, the noncleft side alar base is dragged along but does not move as far, thereby resulting in expansion of the noncleft side nasal base. As the medial footplate pulls away from the noncleft side alar base, the nasal sill contour (white line) changes from an inverted V to one that includes a wide central valley. In addition, the curved noncleft side philtral column (black line) straightens and shortens, presumably because of tension from the muscle repair. On the cleft side, the alar base advances anteriorly, with little to no mediolateral shift, to be united with the centralized columella. Closure of the nasal floor raises the cleft side alar dome. Landmarks along the cleft side nostril rim (yellow dots) allow visualization of the subunits as they change and demonstrate how the cleft side columella increases in height.

Centralization of the columella and anterior advancement of the cleft alar base establishes the lateral alar crural ring³⁴ (Fig. 2). We found that these movements were associated with elevation of the cleft nasal dome and increase in the cleft alar and columellar heights (+2.1 mm and +1.3 mm). In opposition, these changes were associated with subtle but significant narrowing of the columella (−0.2 mm), shortening of the noncleft side columellar height (−0.5 mm), and an observation that the prominence or “boxiness” of the noncleft side hemitip became less pronounced (Fig. 2). Overall, simultaneous lengthening of the cleft side and shortening of the noncleft side columellar heights balanced the ratio from 0.57 to 1.06 (P < 0.0001).

Opposing Lip Changes (P < 0.05)

Closure of lip elements across the cleft resulted in significant elongation of lateral lip height and width on both sides, but with greater changes on the cleft side (+1.6 and +3.4 mm) than on the noncleft side (+0.5 and +2.3 mm) (Fig. 1). As tissues were stretched together, the Cupid’s bow widened (+2.1 mm), while the total lip width (from commissure to commissure) narrowed (–1.6 mm).

While the philtral height ratio normalized for all subjects, we found that this did not occur solely because of elongation on the cleft side. Rather, balance occurred with simultaneous cleft side elongation (+3.0 mm) and noncleft side shortening (−2.1 mm).

Predictors of Noncleft Side Philtral Height Shortening

Table 3 summarizes subjects grouped by the magnitude of noncleft side philtral shortening and the associated preoperative columellar angle, difference in philtral heights, and lateral deviation of subnasale from midline (R = 0.67, R = 0.65, and R = 0.74, respectively).

Table 3. - Noncleft Side Philtral Height Shortening and Associated Preoperative Measurements

Change in Noncleft Side Philtral Height	No. of Patients	Columellar Angle (deg)a	Philtral Height Imbalance (mm)b	sn_x (mm)c
≤1.0 mm	10	19.62 ± 4.33	−3.44 ± 0.39	2.99 ± 0.78
1.1–2.0 mm	15	28.71 ± 2.19	−4.61 ± 0.19	4.82 ± 0.40
2.1–3.0 mm	6	38.34 ± 4.65	−5.34 ± 0.52	6.98 ± 0.75
>3.0 mm	5	44.63 ± 3.39	−5.59 ± 0.42	7.95 ± 0.86
Correlationd		R = 0.67	R = 0.65	R = 0.74

^aColumellar angle is the deviation of columellar axis viewed from submental view from perpendicular to endocanthion–endocanthion (in degrees).

^bPhiltral height imbalance is expressed as noncleft side minus cleft side; philtral heights were measured above the cutaneous roll to match clinical practice and because we recognize the cutaneous roll as a distinct subunit.

^csn_x is the distance of subnasale from endocanthion–endocanthion midline along the x axis (in millimeters).

^dPearson correlation coefficient for preoperative measurement and change in noncleft side philtral height.

Regression analysis revealed that the difference in preoperative philtral heights and the postoperative elongation of cleft philtral height were highly predictive of postoperative shortening (R = 0.85; P < 0.0001). Although the model uses postoperative data and does not facilitate preoperative planning, it suggests that the degree of shortening is related to preoperative severity and the need to elongate the cleft philtral height. To provide an estimate of noncleft side philtral shortening during future repairs, Table 4 summarizes subjects grouped by preoperative measurements and the associated postoperative change.

Table 4. - Preoperative Measurements and Associated Noncleft Side Philtral Height Shortening

Preoperative Measurement	No. of Patients	Change in Noncleft Side Philtral Height (mm)
Columellar anglea
≤15 degrees	4	−0.36 ± 0.21
16–30 degrees	14	−1.13 ± 0.17
31–45 degrees	13	−2.22 ± 0.20
>45 degrees	5	−2.64 ± 0.48
Philtrum height imbalanceb
≤3.0 mm	5	−0.48 ± 0.20
3.1–4.0 mm	5	−0.86 ± 0.40
4.1–5.0 mm	14	−1.78 ± 0.19
>5 mm	12	−2.31 ± 0.28
sn_xc
≤3.0 mm	6	−0.39 ± 0.22
3.1–6.0 mm	16	−1.57 ± 0.17
6.1–9.0 mm	10	−2.21 ± 0.27
>9.0 mm	3	−3.05 ± 0.42

^aColumellar angle is the deviation of columellar axis viewed from submental view from perpendicular to endocanthion–endocanthion (in degrees).

^bPhiltrum height imbalance is expressed as noncleft side minus cleft side; philtral heights were measured above the cutaneous roll to match clinical practice and because we recognize the cutaneous roll as a distinct subunit.

^csn_x is the distance of subnasale from endocanthion–endocanthion midline along the x axis (in millimeters).

Nose Observations

When viewing the nose from a submental perspective, the transition of medial nostril rim from vertical to horizontal along the cleft side nostril is the peak of the “apparent” columellar height (ie, what was measured). The “true” or “anticipated” peak of the columella is found farther along the rim (Fig. 3) and can be visualized using a retractor to elevate the cleft alar dome. When the nostril is not supported, that peak can be anticipated along the medial nostril rim at the transition between concave and convex segments (Figs. 2 and 3).

Fig. 3.: Clinical example is of an Asian patient. The columella is in the nose. (Left) The postoperative columellar height can be anticipated by correcting the position of nasal base elements (blue arrows) and supporting the alar dome. (Center) The apparent height of the columella is at the inflection between the vertical and horizontal contours along the nostril rim. The anticipated height of the columella is farther lateral, can be found along an expanded nostril, and is at the transition where the nostril rim changes from concave to convex. (Right) Correction of the nasal foundation and expansion from caudal septal support (in the absence of any nasal tip dissection) result in elongation of the cleft columellar height, shortening of the noncleft columellar height, and narrowing of the columellar waist.

Opposing subunit shifts occur with repair. On the cleft side, the region that corresponds to the soft triangle,³⁵ or “facet polygon,”³⁶ becomes part of the postoperative columella. On the noncleft side, the columella shortens as columellar skin rotates into the soft triangle (Figs. 2 and 3).

Lip Observations

Although the philtral heights were measured as the absolute distance between crista philtri inferioris and crista philtri superioris, we noticed that the preoperative noncleft philtral column is curved (with a lateral convexity), whereas postoperatively, it tends to straighten (Fig. 2). The column therefore shortens by more than what was measured, and the noncleft side subunit changes shape with surgery.

In addition to noncleft alterations along the philtrum, the upper lip and sill contours also change (Fig. 2). Preoperatively, the noncleft medial footplate abuts the noncleft alar base, resulting in an inverted-V contour of the upper lip along the sill. Postoperatively, constriction of the noncleft alar base width is alleviated, resulting in a distinctive and wide, flat valley between the medial footplate and alar base.

DISCUSSION

Given the refined nature of our craft, it may seem that there is nothing further to discover; however, three-dimensional objective analysis can reveal truths that are hidden in plain sight. Although astute surgeons may have noticed many of these, our findings have been neither discussed nor documented.

Cleft lip presents along a spectrum of presentation in which nasolabial subunits exist in a state of abnormal tension or compression. Following repair, changes (desirable or undesirable) are predicated on how those forces are altered, and clearly, none of the changes occur in isolation because the soft tissues are responding to a multitude of forces and strains, both locally and regionally, before and after repair. Although our goals focus on the cleft side, alterations clearly occur on the noncleft side. Recognition of this is critical given that most surgeons use noncleft side dimensions to guide their surgical plan and assume that the noncleft side represents a static anatomical ideal. If the dynamic nature of the noncleft side is not understood, we will continue to pursue incorrect goals, outcomes will continue to remain suboptimal, and iatrogenic deformities could be produced. Analysis and accurate description of the specific changes that occur on both sides during primary cleft lip repair is therefore a prerequisite to ultimately devising better approaches to treatment.

The Columella Is in the Nose

Mulliken championed the notion that tissue does not need to be added to the columella in the bilateral cleft lip nasal deformity because “the columella only appears small from malpositioned alar cartilages.”^37,38 The same is true for the uCLND. Although we did not add tissue to the columella, the cleft side columellar height elongated with our repair.²⁵

Stenström previously recognized that the short appearance of the cleft side columella is “more apparent than real” because distortion of the columella results in the upper part of it forming a “false nostril rim” and “raising the alar cartilage arch restores the length of the columella.”^11,12 Points marked along the nostril rims before our repairs (Fig. 2) confirm that the additional columellar tissue originates from the alar rim corresponding to the region of the soft triangle. The eventual height of the columella can be anticipated by elevating the nostril rim with a retractor (Fig. 3) or by the transition along the unsupported medial nostril rim between concave and convex contours. Given that we do not perform nasal tip correction,²⁶ elongation seems to occur as balance of the nasal base elements and structural support by means of septoplasty expand the tissues and shift the shape and territory of anatomical subunits.

Although columellar lengthening using the C-flap to add tissue to the columella has been emphasized in many variations of Millard’s repair,^16,39–45 our study demonstrates that there are other ways to balance the columella. C-flap incisions that extend into the columella can be visible, produce additional scar and, in some cases, result in iatrogenic deformity.

Columellar lengthening has also been emphasized as a goal of nasoalveolar molding.⁴⁶ Although presurgical molding is helpful, a focus on tissue expansion and columellar morphology could lead to overexpansion of the nose, resulting in an iatrogenic meganostril. As revealed in our previous audit,⁴⁷ improved convexity of the cleft heminasal tip and reduction of nostril rim recurvatum produce the most favorable outcomes following surgical repair and should be the goals of presurgical molding.

Nasal Sill

With the goal of matching lines of closure to the borders of anatomical subunits following repair, the dynamic nature of the noncleft side needs to be considered. The noncleft nasal sill changes shape through the course of surgery and develops a wide valley in the center of its inverted-V shape (Fig. 2). This valley is part of the upper lip anatomical subunit, which is not apparent before repair. The anatomic subunit approximation accounts for the changing subunits and the eventual shape of the nasal sill because cleft side incisions are based on the relationship of noncleft side landmarks and contours.^23,25

In contrast, Millard’s original repair transposes tissue into the nasal sill and produces a straight horizontal closure. Onizuka et al. recognized that this does not match anatomical subunits and modified their repair so that closure approximates the eventual noncleft side contours.⁴⁸ Surgeons who prefer Millard style repairs should consider the modification of Onizuka et al. or a similar modification.

Philtral Balance

Elongation of the cleft side philtral height is the principal and common goal of all cleft lip repairs.^24,25,39 Shortening of the noncleft side philtral height has yet to be reported, and likely occurs from release of abnormal muscle insertions and tension set by repair (Figs. 4 and 5).

Fig. 4.: Muscle release, primary septoplasty, and nasal floor closure. (Left) On the medial side, orbicularis (yellow) is delaminated from skin and mucosa underneath the columella. With muscle released from alveolus, the lip drops down and a dead space is created (light blue). On the lateral side, muscle (yellow) is delaminated from nasal vestibule and mucosa. With muscle released from alveolus and across to the region farther lateral to the cleft alar base, an empty triangle devoid of muscle is produced (light blue). (Right) Further releases are made so that the alar base can be mobilized anteriorly and sequentially include supraperiosteal dissection along the piriform margin, upper buccal sulcus incision, supraperiosteal dissection of anterior maxilla, and lateral nasal wall advancement flap. The septum (blue) has been released from the anterior nasal spine (cross hatched in black) and secured along the midline of the face. The nasal floor has been closed (white dots) with a lateral nasal wall flap.

Fig. 5.: Muscle repair. The cephalic edge of lateral lip muscle (yellow) is inset into the base of the columella (light blue shading), thereby facilitating columellar centralization and stabilizing the new orientation of the Cupid’s bow. Tissue tension as the medial lip muscle (orange) is inset and balanced with the lateral lip muscle results in flattening, straightening, and shortening of the noncleft philtral column (black). Alteration of the subalare–crista philtri inferioris–crista philtri superioris (sbal-cphi-cphs) triangle (white and black lines) highlights the dynamic nature of the noncleft lateral lip subunit.

Many variations in muscle reconstruction have been reported,^49–55 and the details of our repair have been described.²⁶ Muscle is released from the columellar base to allow the lip to drop, thereby producing dead space (Figs. 4 and 5). On the lateral side, muscle is released from the cleft alar base, with a backcut along the alveolar ridge that produces an empty triangle devoid of muscle lateral to the alar base. Muscle from the lateral lip is inset under the columella such that the tension centralizes the columella. Simple sutures approximate muscle with the goal of aligning the J shape of the orbicularis. We do not operate directly on the noncleft side, and thus, changes there are sequential results that vary with muscle and skin inset and that have implications for the various approaches to cleft lip repair.

The Tennison-Randall technique places a Z-plasty at the center of the philtral deficiency and tends to produce long lips.^56,57 Although differential growth of transposed tissue has been proposed as an explanation, overelongation can be traced to the immediate postoperative result.^57–59 Given that the mechanism of elongation is highly effective, lip height can easily be made excessive, especially when noncleft philtral shortening has not been considered in the design.

Millard style repairs involve a “cut-as-you-go” approach in which surgeons gauge balance as incisions are made. Overelongation of the cleft side is unlikely because of the natural tendency to avoid excessive incisions. However, given that the technique has a propensity to produce short lips, our finding of noncleft philtral shortening should not serve as a reason to “cheat” or to underrotate the philtrum. Rather, our study emphasizes the importance of applying appropriate tension and vectors to mimic the eventual repair when verifying philtral balance. Rather than just downward traction on the philtrum, there should also be a horizontal vector to centralize the lip.

The Fisher technique uses the Rose-Thompson effect^24,60,61 and an inferior triangle, when needed, to elongate the lip. Although lengthening from excision is estimated at 1 mm,^24,25 the actual elongation likely varies with the configuration of the design. When the lip elements are vertically short, as with severe clefts, the limbs of excision become acute. While this produces greater vertical elongation through horizontal recruitment of tissue, the tension likely results in further reduction of the noncleft philtral height. As for the inferior triangle, the base width is determined by measuring the noncleft philtral height at rest and the cleft philtral height with gentle downward traction.^24,25,62,63 Although the latter is intended to account for changes when the lip is unfurled,^24,25 the amount of traction is subjective and could be one way that surgeons anticipate more global alterations of philtral balance. These unaccounted variations could explain why some surgeons have complained of overelongation and Cupid’s bow “blunting,”⁶⁴ while other surgeons have been able to use small triangles with even the most severe of clefts.²⁴ A more objective system of planning that considers the dynamic nature of the noncleft side would therefore facilitate repair. Table 4 provides estimates of noncleft philtral shortening based on preoperative morphology. We are currently using the data in practice to verify our surgical design and are piloting an alternative method of planning that does not require downward traction when measuring the cleft side philtral height. In the meantime, there are three checkpoints subsequent to the lip markings that surgeons should rely on to assess balance and make adjustments during repair.^24,25,62 Regardless of the method of repair, awareness of the mechanics of tissue change can allow us to rely less on artistic judgment and to use data-driven estimates to improve the consistency of our outcomes.

Limitations and Paradigms

Although our observations were visible with each subject, the degree to which these changes occur may vary with individual variations and between different racial groups (Figs. 2, 3, and 6). While the alterations that we have noted likely apply to other methods of repair and other surgeons, further study is needed to confirm this. Although we did not intentionally alter the noncleft side, awareness of the changes has altered how we assess and judge balance during repair.

Fig. 6.: Clinical example of an African American patient. (Left) Preoperatively, the nostril rim was ulcerated because of chronic contact with the lip. There was no apparent cleft side columellar height. (Center) Balance following repair. As the cleft alar dome was raised and cleft columellar height was supported, the noncleft heminasal tip became less pronounced. (Right) Postoperative changes and improved nasolabial balance persist 1 year later.

When insetting muscle into the columella base, we can now see that the lateral muscle is not advancing medially (as the tenet of rotation-advancement would lead us to believe); rather, the tension centralizes the columella while supporting the cleft alar base in a more anterior position. We now look for shortening of the noncleft side philtrum as the prolabium is advanced across the cleft and we can more confidently anticipate columellar heights when designing the points of nasal closure. By identifying these previously unquantified noncleft side alterations, we can be intentional in leveraging anticipated changes to achieve balance. Even if those alterations are variable between surgeons and techniques, the findings emphasize the importance of rechecking balance through the course of repair rather than relying on preoperative measurements that assume the noncleft side is static and normal.

Whereas there is a stigma against altering the noncleft side, that side clearly must change. Beyond the immediate surgical insights, we hope that the findings and data can ultimately stimulate refinement of surgical techniques and approaches to care.

CONCLUSIONS

The uCLND is composed of opposing cleft and noncleft side alterations. Primary repair balances those alterations in opposing ways.

First, centralization of the columella narrows the cleft nasal base and widens the noncleft nasal base. In the absence of any tissue added to the columella, as the cleft columellar height elongates, the noncleft columellar height shortens (and the columellar width narrows). Concomitant correction of the cleft alar base retrusion elevates the cleft alar dome and expands the cleft alar height.

Second, tension from tailoring the lip elements together elongate the cleft and noncleft lateral lip heights and widths, broaden the Cupid’s bow, and reduce total lip width by drawing the commissures closer together. Although repair elongates the cleft philtral height, it significantly reduces the noncleft philtral height. This shortening can be estimated based on preoperative measures of severity.

We observed that the shape, territory, and configuration of cleft and noncleft anatomical subunits change through surgery and that those changes may be anticipated when the interplay of opposing alterations is considered.

Given the results of this study, it is inadequate to simply focus on cleft side corrections. Nasolabial harmony requires balance of the opposing cleft and noncleft side alterations, and we propose that this should be the central goal of reconstruction.

DISCLOSURE

The authors have no financial interest to declare in relation to the content of this article. No funding was received for this work.

PATIENT CONSENT

Parents or guardians provided written informed consent for use of the patients’ images.