Factors associated with alteration of nipple or skin sensation and impact of duration of time following nipple-sparing mastectomy (NSM): an analysis of 460 cases with comparison of conventional versus endoscopic- or robotic-assisted NSM

Compared with skin-sparing mastectomy (SSM), patients who received nipple-sparing mastectomy (NSM) showed significantly improved aesthetic outcomes and quality of life (QoL) [1‐3]. NSM is now increasingly adopted as one of the standards for patients undergoing mastectomy without nipple-areolar complex (NAC) involvement due to acceptable oncologic safety [4‐6].

Diminished or loss of nipple sensation had been reported in 10–75% of patients post-NSM [7‐10]. For this patient group, losing their nipple sensation despite NAC preservation would result in lower satisfaction and QoL than patients who received breast-conserving surgery [11].

Novel surgical techniques of NSM, like endoscopic- or robotic-assisted NSM (R-NSM), had shown a decreased risk of nipple ischemia necrosis [12, 13]. There has been a paucity of studies reporting on the effect of different surgical approaches on nipple or skin sensation post-surgery due to the relatively recent advent of these surgical methods in the past decade. One recently reported randomized controlled trial showed R-NSM to be significantly associated with better nipple sensation compared with conventional nipple-sparing mastectomy (C-NSM) [14]. Further studies or investigations on whether these novel NSM techniques result in better preservation of nipple or skin sensation compared with C-NSM is of utmost importance.

Compared to other research areas of NSM studies, sensation preservation or related factors were underreported. Some studies have reported nipple sensation loss post-NSM, but they were limited to a small number of patients which further limit risk factors analysis of nipple or skin sensation loss [14, 15]. Furthermore, it remains unclear whether the duration of “time since surgery” would result in the recovery of nipple or skin sensation [16, 17].

The aims of the current study were to evaluate the nipple and skin sensation status post-NSM and compared the impact of different surgical approaches, like robotic — versus endoscopic — or conventional NSM on sensation preservation. Further evaluation on the influence of “time since surgery” on the sensation of the nipple or skin and related risk factors would be performed as well.

A total of 460 NSM procedures were enrolled in the current study, and their mean age was 48.3 ± 9.1 years old. There was 383 (83.3%) immediate breast reconstruction, and 77 cases (16.7%) were without breast reconstructions. One-hundred seventy-four (37.8%) received C-NSM, 195 (42.4%) E-NSM, and 91 (19.8%) were R-NSM procedures. The mean body mass index (BMI) was 23.7 ± 3.7, and 116 (27.1%) received radiotherapy after surgery. The clinicopathologic factors were summarized in Table 1.

All NSM procedures with nipple sensation assessed were evaluated at a mean of 42.4 ± 32.5 months post-operation (time since surgery). Fifteen (3.3%) cases were graded as 0, 18.2% (83/460) grade I, 49.7% (229/460) grade II, and 28.9% (133/460) grade III. The overall mean score of nipple sensation was 2.1 ± 0.7, and the average grade score of skin sensation over four quadrants was 2.1 ± 0.7 (Table 2) as well.

In analysis of different surgical approaches, preserved or normal (grade III) nipple sensation rate was 36.2% (63/174) in C-NSM group, 26.7% (52/195) in E-NSM group, and 19.7% (18/91) in the R-NSM group (P = 0.06, Table 2). It has to be noted that “time since surgery to last evaluation” was noted to be significantly longer in the C-NSM group (45.6 ± 34 months) or E-NSM group (44.7 ± 35.8 months) as compared to R-NSM group (31.8 ± 16 months, P < 0.01). (C-NSM group (45.6 ± 34 months) compared with E-NSM group (44.7 ± 35.8 months) was not statistically significant, P = 0.79.) In terms of surgical wound placement, preserved nipple sensation was noted to be 23.9% (17/71) in the upper outer oblique (radial) incision group (time since surgery: 54.6 ± 32.7 months), 38.5% (60/156) in the peri-areolar-related incision group (54.2 ± 37.8 months), 33.3% (6/18) in the inframammary fold incision group (44.8 ± 43.6 months), and 23.3% (50/216) in the single axillary or lateral chest incision (29.63 ± 20.6 months; nipple sensation: P = 0.04, time since surgery: P = 0.01, Table 2).

As for factors related to preserved nipple sensation, 133 (28.9%) grade III nipple sensation NSM procedures group were compared with 327 (71.1%) abnormal nipple sensation group (Table 3). The statistically significant factors for nipple-related sensation are age (50.4 ± 9 versus 47.5 ± 9, P < 0.01), tumor location (P = 0.03), different BMI categories (P = 0.04), different surgical methods (P = 0.01), axillary staging method (P = 0.04), surgical wound location (0.01), and follow-up duration (52.1 ± 32.4 versus 38.4 ± 31.8 months, P < 0.01). Statistically significant differences in skin sensation were found in different surgical methods (P < 0.01), surgical wound locations (P < 0.01), adjuvant radiotherapy (P = 0.01), and follow-up duration (48.5 ± 32.6 versus 39.2 ± 32.1 months, P < 0.01).

Of these 460 NSM procedures, 29 (6.4%) had transient nipple ischemia and 32(6.9%) with partial nipple necrosis. There was no direct correlation between nipple ischemia/necrosis and subsequent nipple sensation. Possible factors of nipple or skin sensation were further validated in univariate and multivariate analysis (Table 4). In univariate analysis, surgical methods, wound location, and follow-up time were noted as significant risk factors. In multivariate analysis, longer follow-up time showed significantly improved nipple or skin sensation (> 60 versus ≦ 12 months: nipple odds ratio (OR) = 5.75, P < 0.01; skin: OR = 1.97, p < 0.05). Peri-areolar wound location showed higher nipple sensation (OR: 2.1, P = 0.02), while single axillary or lateral chest incision showed decreased skin sensation (OR: 0.4, P < 0.01) when compared with the upper outer quadrant incision (Table 4). The R-NSM group showed decreased nipple (OR = 0.4, P < 0.01) or skin (OR = 0.16, P < 0.01) sensation, and E-NSM group also had lower skin sensation (OR = 0.42, P < 0.01) compared with C-NSM group.

The results of the current study were summarized and compared with other reported series in the literature as shown in Table 5.

In this current study, 460 cases NSM procedures from Aug 2011 to April 2022 were enrolled and with clear evaluations of the impact of time lapses on the recovery of nipple or skin sensation. The mean follow-up duration of 42.4 ± 32.5 months (ranged: 1–143) was adequate to evaluate the recovery of skin or nipple sensation. In all cases, the sensation assessment evaluation uses “five areas measurement points” (Fig. 2). With 230 cases having more than one measurement time (some patients even had more than five times evaluated throughout the follow-up), 28.9% (133/460) and 34.5% (159/460) of the NSM procedures retained nipple sensations and skin sensations (grade III), respectively.

Recent studies on the incidence of nipple and skin sensation loss or recovery have varied and inconclusive results due to different maneuvers used and timing of sensation assessment among different reported studies (Table 5). The reported nipple sensation ranged from 10 to 75% (Table 5) [7‐9, 19, 23‐25]. In general, the skin had superior filament discrimination compared to the NAC. A study evaluating 150 healthy women regarding normal breast sensitivity found that the skin of the superior quadrant was the most sensitive part of the breast, followed by the areola, and the least sensitive part was the nipple [26]. Nonetheless, we found that the mean average grade score of breast skin sensation post-NSM over four quadrants was 2.1 ± 0.7, which was similar to the nipple sensation (2.1 ± 0.7, Table 2) in our study.

From univariate and multivariate analysis in our study, factors related to nipple or skin sensation were surgical approaches, wound incision types, and long follow-up duration (Tables 3 and 4). In a recently published RCT [14], R-NSM was associated with better preserved nipple sensation than C-NSM (31.6% versus 0%, P = 0.0002). In the current study, the preserved (grade III) nipple sensation rate was 36.2% (63/174) in the C-NSM group (mean follow up time 45.6 ± 34 months), 26.7% (52/195) in the E-NSM group (44.7 ± 35.8 months), and 19.7% (18/91) in the R-NSM (31.8 ± 16 months) group (nipple sensation: P = 0.06, follow up duration: P < 0.01 (C-NSM or E-NSM versus R-NSM), Table 2). In multivariate analysis, E-NSM or R-NSM showed decreased nipple or skin sensation if compared with C-NSM (Table 4). For patients who received R-NSM or E-NSM (since 2014) procedures, single axillary (or lateral chest incision) was the most frequently adopted incision type, however, which was associated with significantly decreased skin sensation. Peri-areolar incision showed significantly better nipple sensation than upper outer incision (Tables 3 and 4), which was rarely reported before, if any.

“Time from surgery to last evaluation” was a significant factor associated with sensation recovery or preservation in nipple (52.1 ± 32.4 versus 38.4 ± 31.8, P < 0.01) or skin sensation (48.5 ± 32.6 versus 39.2 ± 32.1, P < 0.01) post NSM. In multivariate analysis, follow-up duration is a significant factor in preserving nipple (OR = 5.75, P < 0.01) or skin (OR = 1.97, P < 0.05) sensation. In subgroup analysis of different surgical methods on the impact of nipple or skin sensation, there was significantly longer follow-up duration in C-NSM group (45.6 ± 34 months) or E-NSM (44.7 ± 35.8 months) than R-NSM (31.8 ± 16) group (P < 0.01, Table 2). The concept of “time since surgery” is associated with improved nipple sensation. Rodriguez-Unda et al. [27] suggested that factors such as age, time since surgery, and surgical approach have been linked to nerve regeneration. Shridharani et al. [28] showed that the sensory nerve of the nipple and skin flap “need time” to reinnervation. The actual time needed was unclear; however, it showed progressive improvement of constant touch over time with a lag period of the first 12 months. It seems that it takes 18 to 24 months of sensory recovery after breast reconstruction. These assumptions were consistent with our findings of follow-up time and nipple sensation in Table 4.

The sensory nerves of the NAC and breast skin come from lateral and cutaneous branches of the 3rd–5th intercostal nerves (mainly the 4th lateral cutaneous branch). The lateral cutaneous branches were a greater branch that took a deep course from muscle penetrating breast parenchyma and pierced to the posterior surface of nipple [29‐33]. Montagne and Macpherson et al. [34] demonstrated that the neural elements concentrate at the base of the nipple, with few at the side of the nipple and even fewer in the areolar. Therefore, it is unsurprising how nipple and skin sensations were lost or diminished after NSM [35, 36], especially when retro-areolar tissue sampling was performed for intraoperative frozen section.

In the current study, nipple and skin flap sensation differed according to different surgical incisions and operation methods, which was rarely discussed before. E-NSM or R-NSM did not showed improved nipple or skin sensation than C-NSM even in multivariate analysis (Tables 2, 3 and 4). Skin incision and types of operations were highly correlated. A single axillary or lateral chest incision, which was frequently the preferred incision in E-NSM or R-NSM procedures, and upper-outer, inframammary, or peri-areolar incisions were frequently used by C-NSM procedures (Fig. 3). In multivariate analysis, peri-areolar incision showed significantly higher nipple sensation preservation, while singe axillary or lateral chest incision showed significantly decrease skin sensation (Table 4). Based on the anatomic basis of nerve supply, the single axilla or lateral chest incision, which started dissection from a “lateral to medial” fashion with incision lateral to the lateral border of the breast, might increase the risk of injury of the lateral cutaneous branches of 4th intercostal nerve when coursing into the breast [19]. The dissection plane of peri-areolar, inframammary, or upper outer incision started “medial to lateral” and usually stopped at the lateral border of the breast, which might help in preserving the entry of the lateral cutaneous branch of 4th intercostal nerve and therefore led to higher preservation of skin sensation.

The peri-areolar incision was associated with higher nipple sensation in multivariate analysis (Table 4). In our current study, sub-nipple samplings were routinely performed to prevent occult cancer left behind NAC. Peri-areolar incision was associated with higher risk of NAC ischemia necrosis, and surgeons tend to leave thicker skin flap to avoid over-thinning of retro-areolar tissue and subsequent NAC total necrosis. However, in E-NSM or R-NSM, the risk of total NAC necrosis was decreased due to the placement of skin incision far away from NAC, and usually, the skin flap beneath NAC was thinner due to oncologic safety consideration. These unique technical differences might have contributed to the observation of higher nipple sensation preservation in peri-areolar incision (frequent used in C-NSM) than single axilla or lateral chest incision (usually used in E-NSM or R-NSM, Fig. 3, Tables 3 and 4).

In some studies, there was hypothesis that the factors contributing to nipple necrosis might be correlated with loss of nipple sensation as well owing to the virtue of related neurovascular bundle anatomy when dissecting the superficial breast and skin fascia [20, 21, 37]. However, we found no correlations between NAC ischemia necrosis grading and nipple or skin sensation post-NSM (Table 3). Post-mastectomy radiotherapy (PMRT) was indicated in patients with poorer prognostic factors, like tumor size > 5 cm or lymph nodes metastasis, to decrease disease recurrence and prolonged survival [22]. There was 27% of NSM procedures in the current cohort who received PMRT, but these were not associated with lower preservation of nipple sensation. In contrast, we observed a higher preserved skin sensation in patients who received PMRT (41.3% versus 31.9%, P = 0.01). These findings were consistent with Khan et al. [7], which showed that PMRT did not deteriorate nipple or skin sensation.

In our current study, the focus is on nipple and skin sensation post NSM, and the authors successfully identified factors associated with preserved nipple or skin sensation with further validation of the relationship between nipple or skin sensation recovery to the duration of “time since surgery.” Our study is limited due to its retrospective nature, small sample size, and possible selection bias. Furthermore, the results of nipple or skin sensation test were subjective and not equally measured within a specific time period. However, our study enrolled a large number of NSM cases with detailed clinicopathologic, perioperative parameters, and different surgical approaches or incisions to enable us to have a comprehensive analysis of possible factors related to preserved nipple or skin sensation post NSM. A long-term follow-up of mean 42.4 ± 32.5 months also enabled us to show gradual recovery of nipple or skin sensation with different “time after surgery” period. Therefore, the results and information derived from the current study are valuable.

In our current study, the authors successfully demonstrated strong evidence that a significant improvement of nipple and skin sensation in patients who received NSM had to do with a longer “time after surgery” period. The results derived from this study will thus enable better discussion with patient regarding the impact of different operations methods and skin incision on skin and nipple sensation. This is definitely valuable in improving patients’ outcomes and QoL following NSM, especially in an era of shared decision-making prior to surgery.