European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180

https://doi.org/10.1007/s00405-020-06511-9

REVIEW ARTICLE

Punch vs open surgical techniques for placement of bone‑anchored

hearing implants: a systematic review and meta‑analysis of skin
reactions and operating time
Qianyu Xiao1 · Fanghua Gong1 · Ning Wang2 · Weihua Hu1

Received: 14 September 2020 / Accepted: 19 November 2020 / Published online: 2 January 2021
© Springer-Verlag GmbH Germany, part of Springer Nature 2021

Abstract
Purpose Several authors have reported their experience with the punch technique as compared to open surgical methods
for bone-anchored hearing implants (BAHI). However, no study has attempted to aggregate current evidence. We aimed to
compare post-operative skin complications and operating time between punch and open surgical techniques of BAHI via a
systematic review and meta-analysis.
Methods Databases of PubMed, Embase, Scopus, BioMed Central, Ovoid, and CENTRAL were screened up to 15th Febru-
ary 2020 to include studies comparing punch and open surgical technique for BAHI.
Results Eight studies were included. Punch technique was compared with dermatome and linear incision techniques with
and without soft tissue reduction. There was no difference in normal-to-moderate skin reaction between the punch and open
surgical techniques (OR: 0.86 95% CI 0.23, 3.28 I2 = 0%). The incidence of adverse skin reactions were also not different
between the two groups. Meta-regression for different follow-up periods did not demonstrate any statistically significant
results. Our results also indicated that punch technique requires less operating time, however, the inter-study heterogeneity
in the analysis was very high. Similar results were seen on sub-group analysis based on the type of open surgical technique.
Conclusion There may be no difference in skin tolerance between the punch technique and open surgical techniques. Oper-
ating time may be significantly reduced with the punch technique. Strong conclusions cannot be drawn owing to a limited
number of studies. Further large-scale randomized trials are required.

Keywords Minimally invasive surgery · Complications · Punch technique · Hearing implants · Operating time

Introduction loss like conductive or mixed hearing loss or single-sided

deafness. It has been an acceptable choice of treatment in
Since their first description in the 1970s, osseointegrated subjects not benefiting from conventional therapies, such
bone-anchored hearing implants (BAHIs) have successfully as air conduction hearing aids or reconstructive middle ear
rehabilitated patients suffering from various types of hearing surgery [1–3]. The device consists of an external sound pro-
cessor, which is attached to a titanium implant placed into
the temporoparietal skull region behind the ear. The sys-
Supplementary Information The online version contains tems process the sound and convert into vibrations which are
supplementary material available at https​://doi.org/10.1007/s0040​ transmitting via bone conduction directly to the cochlea [4].
5-020-06511​-9.
Concerns associated with the use of BAHI include skin-
* Qianyu Xiao related complications such as periabutment inflammation,
zhangp_123@126.com skin overgrowth, numbness, pain, and implant extrusion [5].
1
The skin tolerance around the implant is usually measured
Department of Nursing, Hunan Provincial People’s Hospital, on the five-point Holger’s scale and scores of ≥ 2 are fre-
the First-Affiliated Hospital of Hunan Normal University, 61
Jiefang West Road, Changsha, Hunan, China quently associated with discomfort in BAHI patients [6, 7].
2 These adverse events have been partly attributed to the inva-
Department of ENT & HN Surgery, Hunan Provincial
People’s Hospital, the First-Affiliated Hospital of Hunan sive nature of surgery for the placement of BAHI. Authors
Normal University, Changsha, China have hypothesized that invasive methods like the dermatome

13
Vol.:(0123456789)
3172 European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180

technique or the linear incision techniques traumatize the search strategy along with the results of the PubMed data-
skin around the implant and less invasive methods may base are presented in Supplementary Table 1. The review-
improve outcomes [8, 9]. In an attempt to reduce soft-tissue ers screened the search results initially by their titles and
complications, a minimally invasive punch technique has abstracts for every database. After identifying potentially
been developed in recent years. In the ’punch-only’ tech- pertinent articles, full texts of the articles were sourced.
nique, the implant is placed through a key-hole thereby lim- Both the reviewers assessed individual articles based on the
iting soft-tissue damage. The procedure is thought to reduce inclusion and exclusion criteria. Any disagreements were
operating time, improve soft tissue healing, and lead to bet- resolved by discussion. Post-screening, the bibliography of
ter postoperative outcomes [10, 11]. On the downside, the included studies, as well as review articles on the subject,
technique offers limited intraoperative visibility which may were hand searched for any additional references.
lead to damage to subcutaneous tissues.
Over the years, several authors have reported their experi- Data extraction and risk of bias assessment
ence with the punch technique in reducing soft-tissue com-
plications and operating time for BAHI [12–14]. However, Two reviewers independently extracted data from the
to date, no study has attempted to systematically review the included studies. Data regarding authors, publication year,
available evidence. Therefore, this purpose of this study was study type, sample size, demographic details, surgical
to conduct a systematic review and meta-analysis evaluat- details, abutment length, implant system, outcomes, and fol-
ing the role of punch technique in reducing post-operative low-up were extracted. The primary outcome of our analysis
skin complications and operating time as compared to open was skin tolerance assessed using Holger’s classification [6].
surgical techniques for placement of BAHI. Scores of 0–1 were classified as normal-to-moderate skin
reaction whereas scores of ≥ 2 were classified as an adverse
skin reaction. The secondary outcomes of interest were oper-
Materials and methods ating time, the incidence of intra-operative complications,
and revision surgeries.
Inclusion criteria For RCTs, the Cochrane Collaboration risk assessment
tool was used to assess the quality of included studies [16].
This systematic review and meta-analysis were conducted Studies were assessed for: random sequence generation,
as per the PRISMA statement (Preferred Reporting Items allocation concealment, blinding of participants and person-
for Systematic Reviews and Meta-analyses) [15] and guide- nel, blinding of outcome assessment, incomplete outcome
lines of the Cochrane collaboration [16]. We used the PICOS data, selective reporting, and other bias. For non-RCTs, the
(Population, Intervention, Comparison, Outcome, and Study risk of a bias assessment tool for non-randomized studies
design) model to select studies for inclusion in the review. (RoBANS) was used [17]. Studies were assessed for: selec-
All types of studies conducted on patients undergoing BAHI tion of participants, confounding variables, intervention
surgery (Population) were considered for inclusion. Studies measurements, blinding of outcome assessment, incomplete
were to compare the minimally invasive punch technique outcome data, and selective outcome reporting.
without any additional skin incision or soft tissue reduction
(Intervention) with any other incision or flap technique with Statistical analysis
or without soft tissue reduction (Comparison) and measur-
ing operating time, skin tolerance, or other complications as The software “Open MetaAnalyst” was used for the meta-
Outcomes. Studies were included irrespective of sample size analysis [18]. Predicting heterogeneity in the included stud-
and language of publication. Studies not reporting relevant ies, we used a random-effects model to calculate the pooled
data, single-arm studies, review articles, and case reports effect size for all analyses. We assessed heterogeneity using
were excluded. In the case of duplication of data, the most the ­I2 statistic. I­ 2 values of 25–50% represented low, values
updated version of the study was to be included. of 50–75% medium, and more than 75% represented sub-
stantial heterogeneity. Continuous data were summarized
Search strategy using the mean difference (MD) with 95% confidence inter-
vals (CI) and categorical data using Mantel–Haenszel Odds
An electronic search was performed independently by two Ratio (OR) with 95% CI. Sub-group analysis was performed
reviewers. Databases of PubMed, Embase, Scopus, BioMed for the type of comparative technique. We also assessed the
Central, Ovoid, and CENTRAL (Cochrane Central Regis- influence of final follow-up time on the effect size of the
ter of Controlled Trials) were screened up to 15th Febru- primary outcome using random-effects meta-regression
ary 2020. Both MeSH terms and free-text keywords were analysis. For studies not reporting mean and standard devia-
utilized for searching relevant articles. Search terms and tion scores of operating time, the same was calculated from

13
European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180 3173

available data using methods reported by Wan et al. [19]. In Details of the included studies are presented in Table 1.
case, data were represented only graphically, the software Only one study [23] was an RCT while remaining were
Engauge Digitizer was used to extract numerical data. Due retrospective or prospective studies. All studies were con-
to the inclusion of fewer than ten studies in the review, fun- ducted in North America or European nations. The sam-
nel plots were not used to assess publication bias. ple size of the punch technique group varied from 10 to
65 in the included studies, while that of the open surgical
group varied from 8 to 155. In three studies, pediatric
Results patients were also included [11, 22, 25]. Four studies [11,
20–22] used a ‘skin punch’ for the punch technique while
The study flowchart is presented in Fig. 1. Fourteen studies the remaining four studies used the ‘Minimally Invasive
were analyzed by their full-texts of which six were excluded Ponto Surgery’ (MIPS) system for the same [23–26]. The
as they did not fulfill the inclusion criteria. A total of eight open surgical technique varied amongst included studies.
studies were included in this systematic review and meta- The dermatome technique was used in two studies [20,
analysis [11, 20–26]. 21] while the remaining used the linear incision technique

Fig. 1  Study flowchart

13
 3174

Table 1  Characteristics of included studies

13
Author/year Study type Country Sample size Mean Age Abutment length Punch technique Comparative tech- Implant system Follow-up
nique
PT CT PT CT PT CT

Wilson et al. Retro-spective USA 29 11 61.9 70.5 NA NA Using a 4 mm Using a dermatome NA At least 12 m
[20]/2013 diameter biopsy with soft tissue
punch with a reduction
blade length of
6 mm
Gordon et al. Retro-spective USA 17 34 52.2 49.1 9 mm 6 mm Using a 6 mm skin Linear incision BAHA® and PT: mean 10 m
[11]/2015 biopsy punch with soft tissue PONTO® CT: mean 22 m
reduction
Dumon et al. Prospective France 65 155 58a 57a 9 mm 6 mm Using a soft tissue Using a dermatome 150 BAHA® and PT: median 8 m
[21]/2017 (6–12)a punch with soft tissue 70 PONTO® CT: median 21 m
reduction
Bonilla et al. Retro-spective Spain 18 16 52a 58a 10 mm 6 mm Using a 6 mm skin Linear incision BAHA® in all Mean 24 m
[22]/2017 (8–12) (6–12) punch with soft tissue patients
reduction
Calon et al. RCT​ Netherlands 33 30 50.3 51.9 9 mm: 21 9 mm: Using the Mini- Linear incision PONTO® in all 12 weeks
[23]/2018 12 mm: 13 mally Invasive without soft tis- patients
10 12 mm: Ponto Surgery sue reduction
14 mm: 2 16 system
14 mm:
1
Giustino et al. Retro-spective Italy 10 G1: 12 60.41 NA NA Using the Mini- G1: Linear incision PT and G1: 12 months
[24]/2018 G2: 8 mally Invasive with soft tissue BAHA®
Ponto Surgery reduction G2: PONTO®
system G2: Linear incision
without soft tis-
sue reduction
Bezdijian et al. Retro-spective Canada 21 38 40.43 16 NA NA Using the Mini- Linear incision PONTO® in all Up to 15 weeks
[25]/2020 mally Invasive without soft tis- patients
Ponto Surgery sue reduction
system
Caspers et al. Prospective Netherlands 25 25 60 52 6 mm: 1 6 mm: 0 Using the Mini- Linear incision PONTO® in all 6 months
[26]/2020 with historical 9 mm: 14 9 mm: mally Invasive without soft tis- patients
control 12 mm: 17 Ponto Surgery sue reduction
10 12 mm: system
8

PT punch technique, CT comparative technique, NA not available, G1 group 1, G2 group 2

a
Median with or without (range)
European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180
European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180 3175

with or without soft tissue reduction. Both BAHA® and Outcomes

PONTO® implant systems were used in the included
studies. The final follow-up varied across studies from Data on skin tolerance based on Holger’s score at the final
3 to 24 months. follow-up in the included studies are presented in Table 2.
Two studies [21, 25] did not report data on skin tolerance at
the final follow-up, of which one study [25] was not included
in the meta-analysis of the primary outcome. A previous

Table 2  Details of complications in the included studies

Study Holgers classification at final Intra-operative complications Revision surgery and Numbness and dehiscence
follow-up implant loss
Group 0 1 2 3 4

Wilson et al. [20] PT 24 1 3 1 0 NR Revision surgeries- NR

CT 9 0 1 1 0 PT: Soft tissue reduction (1)
LI: Nil
Implant loss-Nil
Dumon et al. [21]a PT 12 2 3 0 0 PT: Bleeding (2), Implanta- Revision surgeries- NR
CT 17 1 1 0 1 tion not possible through PT: Osseointegration failure
the limited punch skin (2), Traumatic implant loss
resection due to interposi- (2), Abutment too long (2)
tion of periosteum between CT: Skin overgrowth (18),
the implant and the bone Non-user (4), Pain (3),
(2) Poor skin tolerance (2),
CT: 3 Bleeding (2), CSF Skin flap necrosis (1),
leak (1) Poor hearing tolerance (1),
Osseointegration failure (1)
Implant loss-
PT: 4
CT: 1
Gordon et al. [11] PT 13 2 1 1 0 NR Revision surgeries- NR
CT 19 12 1 1 0 NR
Implant loss-
PT: 1
Bonilla et al. [22] PT 16 1 1 0 0 PT: Nil NR NR
CT 13 1 2 0 0 CT: CSF leakage (1)
Calon et al. [23] PT 18 14 0 1 0 PT: Drilling into vein (2), Revision surgeries- No significant difference in
CT 18 10 2 0 0 Bleeding hematoma (1) NR skin dehiscence between
CT: Drilling into vein (2), Implant loss- the two groups. Loss of
Dura mater exposed (1), PT: 4 sensibility was sig-
Bleeding (1) CT: 1 nificantly less PT group
compared to CT group
Giustino et al. [24] PT 10 0 0 0 0 PT: Dura mater exposed with NR NR
CT G1 10 0 0 0 1 bleeding (1)
CT G2 7 1 0 0 0 G1: Dura mater exposed
with bleeding (1), Bleeding
(2)
G2: Nil
Bezdijian et al. [25] NR for final follow-up NR NR NR
Casper et al. [26] PT 23 2 Nil Revision surgeries- Increased minimal skin
CT 18 7 Nil dehiscence in PT groups
Implant loss- at 1 and 3 weeks. No dif-
PT-3 ference in skin dehiscence
CT-0 and sensibility between
the two groups at long-
term follow-up

Numbers indicate number of cases

PT punch technique, CT comparative technique, NR not reported, G1 group 1, G2 group 2, CSF cerebrospinal fluid
a
Data from Dumon et al. [13] 2016 for skin tolerance

13
3176 European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180

publication of Dumon et al. [13] with the overlapping sam-

ple was used to extract data of skin tolerance of 37 patients.
The incidence of adverse skin reactions was 8.7% (13/149
patients) with the punch technique, 12.9% with the der-
matome technique (4/31 patients), 8.06% (5/62 patients)
in the linear incision with soft tissue reduction group, and
14.28% (9/63 patients) in the linear incision without soft
tissue reduction group. Based on the type of open surgical

Fig. 2  Meta-analysis of normal-to-moderate skin reaction between punch vs open surgical techniques with sub-group analysis based on type of open surgical technique
technique used, we conducted a sub-group analysis for skin
tolerance and operating time. For normal-to-moderate skin
reactions (Holger’s score 0–1), our meta-analysis indicated
no significant difference between punch technique vs der-
matome technique with soft tissue reduction (OR:0.86 95%
CI 0.23, 3.28 I2 = 0%), punch technique vs linear incision
technique with soft tissue reduction (OR:1.49 95% CI 0.38,
5.80 I2 = 0%) and punch technique vs linear incision tech-
nique without soft tissue reduction (OR:3.22 95% CI 0.86,
11.97 I2 = 0%) (Fig. 2). In the overall analysis, there was no
statistically significant difference in the incidence of normal-
to-moderate skin reactions between the punch technique and
the open surgical techniques (OR: 1.62 95% CI 0.75, 3.50
I2 = 0%).
Similarly, for adverse skin reactions (Holger’s score ≥ 2),
our analysis indicated no statistically significant difference
between punch technique vs dermatome technique with
soft tissue reduction (OR: 1.16 95% CI 0.30, 4.41 I2 = 0%),
punch technique vs linear incision technique with soft tissue
reduction (OR: 0.90 95% CI 0.21, 3.74 I2 = 0%) and punch
technique vs linear incision technique without soft tissue
reduction (OR: 0.31 95% CI 0.08, 1.16 I2 = 0%) (Fig. 3).
In the overall analysis, there was no statistically significant
difference in the incidence of adverse skin reactions between
the punch technique and the open surgical techniques (OR:
0.67 95% CI 0.31, 1.47 I2 = 0%).
As the skin tolerance scores were recorded at different
follow-up time points amongst the included studies, we con-
ducted a meta-regression analysis to assess any influence
of this moderator on the overall effect size. Our analysis
demonstrated no statistical significant influence of follow-up
time on normal-to-moderate skin reactions (ß: -0.050 95%
CI -0.281, 0.180 p = 0.669) (Supplemental Fig. 1) as well as
on adverse skin reactions (ß: -0.076 95% CI -0.157, 0.309
p = 0.523) (Supplemental Fig. 2).
Limited data were available for secondary outcomes,
intra-operative complications, revision surgery, implant loss,
skin numbness, and dehiscence. Therefore, a meta-analysis
was not conducted. Details of these complications are pre-
sented in Table 2. Data on operating time were reported
by all eight studies. Pooled analysis indicated significantly
reduced operating time with punch technique as compared
to dermatome technique with soft tissue reduction (MD:
− 21.41 95% CI − 23.55, − 19.28 I2 = 0%), compared to
linear incision technique with soft tissue reduction (MD:

13
European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180 3177

− 38.77 95% CI − 44.93, − 32.61 I2 = 32.9%) and compared

to linear incision technique without soft tissue reduction
(MD: − 17.62 95% CI − 25.21, − 10.02 I2 = 96%) (Fig. 4).
Overall, the punch technique required significantly reduced
surgical time as compared to open surgical techniques (MD:
− 24.76 95% CI − 31.94, − 17.59 I2 = 97%), but with high
inter-study heterogeneity in the pooled analysis.

Quality of included studies

The author’s judgment of the risk of bias in included studies

Fig. 3  Meta-analysis of adverse skin reaction between punch vs open surgical techniques with sub-group analysis based on type of open surgical technique
is presented in Supplemental Table 2. Owing to the study
design blinding was not possible in the RCT [23]. The RCT
was sponsored by the manufacturer of the MIPS system. For
non-RCTs, the overall quality was not high.

Discussion

Our review indicates that, to date, the punch technique of

BAHI has been compared with three different open surgi-
cal techniques, namely, the dermatome technique, the linear
incision technique with soft tissue reduction, and the linear
incision technique without soft tissue reduction. On pooled
analysis of a limited number of studies with a small sample
size, we found that there is no statistically significant differ-
ence in skin tolerance scores between the punch technique
and other invasive open surgical techniques. However, the
operating time is significantly reduced with the use of the
punch technique.
Surgical techniques of BAHI have gradually evolved over
the past decades with a sole aim to simplify the procedure
and reduce complications. The standard surgical technique
for many years was the skin-flap technique. It involved the
creation of a U-shaped pedicled skin flap with underlying
soft-tissue reduction. In recent times, to standardize the
procedure and reduce the thickness of the flap, surgeons
recommend the use of a dermatome to elevate the skin
flap (the dermatome technique) [27, 28]. Once the flap is
elevated, the subcutaneous tissue is thinned with a gradual
slope towards the implant site. In our review, two studies
utilized this technique in the control group. Another popular
technique is the linear incision method wherein a 30 mm
longitudinal incision is made in the retro-auricular region.
After exposure, the periosteum is mobilized and an implant
placed. Subsequently, subcutaneous tissue is thinned in an
area of 2 cm around the implant site [7, 29]. In our review,
three studies used this technique for comparison with the
punch technique. The rationale of soft-tissue reduction in
the above-mentioned techniques was the hypothesis that
skin movement around the implant site caused adverse tissue
reactions [23]. Sub-cutaneous tissue reduction was therefore
carried out to minimize tissue movement.

13
3178 European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180

Fig. 4  Meta-analysis of operating time between punch vs open surgical techniques with sub-group analysis based on type of open surgical tech-
nique

This has been challenged by Van De Berg et al. [9] who skin reactions are due to the implanted foreign body and the
have suggested that less invasive the surgery, better are the length of the surgical incision and the associated surgical
outcomes. Following this principle and aided with the avail- trauma may have little role in the inflammatory reaction. It
ability of longer implant abutments, the linear incision tech- is also important to note that the outcomes were reported at
nique without soft tissue reduction has gained popularity different follow-up times in the included studies. Therefore,
[30]. Hogsbro et al. [31] in an RCT have shown that the lin- a meta-regression analysis was performed which, however,
ear incision technique without soft tissue reduction results in did not demonstrate any statistical significance.
faster healing, less post-operative pain, and sensibility loss Skin reactions after BAHI surgery can be influenced by
as compared to the dermatome technique. However, long- the post-operative protocol which includes use of antibiotics,
term skin tolerance was similar in the two groups. Despite appropriate wound care and regular follow-up. Although,
being less invasive than the dermatome technique, this linear in a recent retrospective case–controlled study, Kazi et al.
incision technique still requires elevation of the skin flap [32] have reported that the use of post-operative antibiotics
with soft-tissue dissection and suturing at the end of the may not reduce the incidence of skin reactivity in patients
procedure. To further minimize soft tissue injury, the punch undergoing MIPS surgery. The authors suggested that skin
technique was described by Goldman et al. in 2013 [10]. reactions, when occur, are not mediated by infection and
Since the implant is placed through a key-hole without any are directed towards the foreign body. Furthermore, studies
soft tissue reflection and tissue reduction, skin complications have also stressed on the importance of patient education
were thought to reduce with this method. to maintain implant hygiene especially in the early post-
On analysis of skin tolerance scores using the Holger’s operative period and the need of regular follow-up appoint-
classification, we found that there was no significant dif- ments to reduce skin complications [33]. However, due to
ference between the punch technique and either of the unavailability of data, we could not assess the impact of such
three open surgical techniques. The incidence of adverse variables on the outcomes of this review.
skin reactions in 149 cases implanted with the punch tech- Other important adverse skin complications with BAHI
nique was 8.7%. On the other hand, adverse skin reactions surgery include skin numbness and dehiscence. Due to the
were seen in 12.9% cases in the dermatome group, 8.06% key-hole incision of punch technique, less neural damage
cases in the linear incision with soft tissue reduction group, may be expected and therefore reduced skin numbness
and 14.28% cases in the linear incision without soft tissue [26]. However, we could not find strong evidence on the
reduction group. In the absence of an incision in the punch influence of the punch technique on these complications.
group, the risk of adverse skin reactions is expected to be Only two studies [23, 26] compared skin numbness and
low. Lack of statistical significance in our results can be dehiscence between the punch group and linear incision
attributed to two reasons. First, in view of the small sample with the soft tissue reduction group. On long-term follow-
size and a limited number of studies in the review, the meta- up, both did not report any significant difference in the
analysis may not have been adequately powered to detect rates of skin dehiscence around the implant sites. In the
significant differences. Second, it may be argued that the RCT of Calon et al. [23], skin sensibility was significantly

13
European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180 3179

better in the punch group but in the study of Casper et al. Nevertheless, this study is the first review to systemati-
[26], no difference was detected in the two cohorts after a cally synthesize evidence on the outcomes of the punch tech-
follow-up of 6 months. nique as compared to open surgical techniques. We pooled
Our analysis did demonstrate a statistically significant together the dermatome and linear incision techniques
reduction of operating time with the use of the punch tech- against the punch technique as they all invasive surgical pro-
nique. The reduced surgical time may be beneficial in terms cedures as compared to the punch technique where no inci-
of reducing hospital costs of BAHI surgery. Sardiwalla et al. sion is involved. Appropriate sub-group analysis was carried
[34] in their comparison of punch technique vs open surgi- out for the different comparative open surgical techniques
cal techniques have shown that the use of punch technique in our analysis to provide clarity on the current evidence.
reduces hospital costs by a mean of 456.83 Canadian Dol- We also performed a meta-regression analysis to factor in
lars. The procedure may be performed in an office setting the difference of follow-up periods in the included studies.
under local anesthesia thereby eliminating anesthesia and To conclude, our review indicates that there may be no
admission costs [35]. It is important to note that the overall difference in skin tolerance between the punch technique
pooled analysis for operating time had high inter-study het- and open surgical techniques. Operating time is significantly
erogeneity. This may be due to the different open techniques reduced with the punch technique as compared to all open
pooled together for comparison with the punch technique. surgical methods. However, strong conclusions cannot be
Disadvantages of the punch technique include the limited drawn owing to a limited number of studies in this review.
visibility of the surgical field which may lead to damage to Also, data regarding intra-operative and other skin complica-
soft tissues, difficulty in managing intra-operative complica- tions are, at present, limited. Further large-scale RCTs are
tions like bleeding, and non-perpendicular insertion of the needed to strengthen the evidence on this subject.
implant in the bone [25]. The absence of adequate irrigation
may also lead to high bone temperatures causing thermal
osteonecrosis thereby affecting implant stability [25, 26]. On Authors’ contributions QX conceived and designed the study. FG, NW
and WH collected the data and performed the literature search. QX was
descriptive analysis of data, the incidence of intra-operative involved in the writing of the manuscript. All authors have read and
complications and implant loss was low in all included stud- approved the final manuscript.
ies with no statistically significant difference between punch
technique and other open surgical techniques. Furthermore, Funding None.
not all studies reported data on these outcomes, which fur-
ther limits our ability to draw conclusions. Availability of data and materials The datasets used and/or analyzed
during the current study are available from the corresponding author
The limitations of our review need to be specified. First, on reasonable request.
only eight studies were available for inclusion in the review.
Also, the punch technique was compared with different open Compliance with ethical standards
surgical techniques in the included studies. The number of
studies in each comparative sub-group was not more than Conflicts of interest None to declare.
three. The sample size of the included studies was small
and despite pooling results, our analysis may not have been
adequately powered. Also, not all outcome variables were
studied by all included studies. Second, only one study was References
an RCT while all others were non-randomized studies with
a high risk of bias. This may have skewed the results of 1. Tjellstrom A, Lindstrom J, Hallen O et al (1981) Osseointegrated
our analysis. Third, there were methodological variations in titanium implants in the temporal bone. A clinical study on bone-
anchored hearing aids. Am J Otol 2:304–310
the included studies for the instruments used for the punch 2. Edmiston RC, Aggarwal R, Green KMJ (2015) Bone conduc-
technique, the implant system, and the length of abutments. tion implants—a rapidly developing field. J Laryngol Otol
The influence of these variables on the overall results is not 129:936–940
known. Fourth, patient co-morbidities, post-operative anti- 3. Crowson MG, Tucci DL (2016) Mini review of the cost-effec-
tiveness of unilateral osseointegrated implants in adults: possibly
biotic protocol, and post-operative care can also affect the cost-effective for the correct indication. Audiol Neurotol 21:69–71
incidence of complications. These data were, however, not 4. Stenfelt S, Goode RL (2005) Bone-conducted sound: physiologi-
available from the included studies. Lastly, outcomes of our cal and clinical aspects. Otol Neurotol 26:1245–1261. https​://doi.
review may have been influenced by several factors like the org/10.1097/01.mao.00001​87236​.10842​.d5
5. Kiringoda R, Lustig LR (2013) A meta-analysis of the complica-
experience of the operating surgeon, geographical variations tions associated with osseointegrated hearing aids. Otol Neurotol
in the surgical protocol, the ethnicity of the patient popula- 34:790–794. https​://doi.org/10.1097/MAO.0b013​e3182​91c65​1
tion, etc. These factors should be taken into account while 6. Holgers KM, Tjellstrom A, Bjursten LM, Erlandsson BE (1988)
interpreting our results. Soft tissue reactions around percutaneous implants: a clinical

13
3180 European Archives of Oto-Rhino-Laryngology (2021) 278:3171–3180

study of soft tissue conditions around skin-penetrating titanium a retrospective single center comparative study. Acta Otorrino-
implants for bone-anchored hearing aids. Am J Otol 9:56–59 laringol Espanola 68:309–316. https​://doi.org/10.1016/j.otorr​
7. Dun CAJ, Faber HT, De Wolf MJF et al (2012) Assessment of i.2017.01.005
more than 1,000 implanted percutaneous bone conduction devices: 23. Calon TGA, Johansson ML, De Bruijn AJG et al (2018) Mini-
skin reactions and implant survival. Otol Neurotol 33:192–198. mally invasive ponto surgery versus the linear incision tech-
https​://doi.org/10.1097/MAO.0b013​e3182​41c0b​f nique with soft tissue preservation for bone conduction hearing
8. De Wolf MJF, Hol MKS, Huygen PLM et al (2008) Clinical out- implants: a multicenter randomized controlled trial. Otol Neurotol
come of the simplified surgical technique for BAHA implantation. 39:882–893. https​://doi.org/10.1097/MAO.00000​00000​00185​2
Otol Neurotol 29:1100–1108. https:​ //doi.org/10.1097/MAO.0b013​ 24. Di Giustino F, Vannucchi P, Pecci R et al (2018) Bone-anchored
e3181​8599b​8 hearing implant surgery: our experience with linear incision and
9. Van De Berg R, Stokroos RJ, Hof JR, Chenault MN (2010) Bone- punch techniques. Acta Otorhinolaryngol Italica 38:257–263.
anchored hearing aid: a comparison of surgical techniques. Otol https​://doi.org/10.14639​/0392-100X-1694
Neurotol 31:129–135. https​://doi.org/10.1097/MAO.0b013​e3181​ 25. Bezdjian A, Smith RA, Gabra N et al (2020) Experience with
c29fe​c minimally invasive ponto surgery and linear incision approach
10. Goldman RA, Georgolios A, Shaia WT (2013) The punch method for pediatric and adult bone anchored hearing implants. Ann Otol
for bone-anchored hearing aid placement. Otolaryngology—head Rhinol Laryngol 129:380–387. https​://doi.org/10.1177/00034​
and neck surgery (United States). Otolaryngol Head Neck Surg 89419​89145​1
12:878–880 26. Caspers CJI, Kruyt IJ, Mylanus EAM, Hol MKS (2020) Six-
11. Gordon SA, Coelho DH (2015) Minimally invasive surgery for month clinical outcomes for bone-anchored hearing implants:
osseointegrated auditory implants: a comparison of linear ver- comparison between minimally invasive ponto surgery and the
sus punch techniques. Otolaryngology - Head and Neck Surgery linear incision technique with tissue preservation. Otol Neurotol
(United States) 152:1089–1093. https​://doi.org/10.1177/01945​ 41:e475–e483. https:​ //doi.org/10.1097/MAO.000000​ 00000​ 02562​
99815​57153​2 27. Stalfors J, Tjellström A (2008) Skin reactions after BAHA sur-
12. Johansson ML, Stokroos RJ, Banga R et al (2017) Short-term gery: a comparison between the U-graft technique and the BAHA
results from seventy-six patients receiving a bone-anchored hear- dermatome. Otology and Neurotology 29:1109–1114. https​://doi.
ing implant installed with a novel minimally invasive surgery tech- org/10.1097/MAO.0b013​e3181​85fab​c
nique. Clin Otolaryngol 42:1043–1048 28. Strijbos RM, Bom SJH, Zwerver S, Hol MKS (2017) Percutaneous
13. Dumon T, Medina M, Sperling NM (2016) Punch and drill: bone-anchored hearing implant surgery: dermatome versus linear
Implantation of bone anchored hearing device through a minimal incision technique. Eur Arch Otorhinolaryngol 274:109–117. https​
skin punch incision versus implantation with dermatome and soft ://doi.org/10.1007/s0040​5-016-4210-3
tissue reduction. Ann Otol Rhinol Laryngol 125:199–206. https:​ // 29. Mohamad S, Khan I, Hey SY, Hussain SSM (2016) A system-
doi.org/10.1177/00034​89415​60644​7 atic review on skin complications of bone-anchored hearing aids
14. Sardiwalla Y, Jufas N, Morris DP (2018) Long term follow-up in relation to surgical techniques. Eur Arch Otorhinolaryngol
demonstrating stability and patient satisfaction of minimally inva- 273:559–565
sive punch technique for percutaneous bone anchored hearing 30. Husseman J, Szudek J, Monksfield P et al (2013) Simplified bone-
devices. J Otolaryngol Head Neck Surg. https​://doi.org/10.1186/ anchored hearing aid insertion using a linear incision without
s4046​3-018-0316-5 soft tissue reduction. J Laryngol Otol 127:S33–S38. https​://doi.
15. Moher D, Liberati A, Tetzlaff J et al (2009) Preferred reporting org/10.1017/s0022​21511​30007​41
items for systematic reviews and meta-analyses: the PRISMA 31. Høgsbro M, Agger A, Johansen LV (2015) Bone-anchored hearing
statement. PLoS Med 6:e1000097. https​://doi.org/10.1371/journ​ implant surgery: randomized trial of dermatome versus linear inci-
al.pmed.10000​97 sion without soft tissue reduction-clinical measures. Otol Neurotol
16. Higgins J, Thomas J, Chandler J, et al (2019) Cochrane handbook 36:805–11. https​://doi.org/10.1097/MAO.00000​00000​00073​1
for systematic reviews of interventions, Version 6. Cochrane 32. Kazi AA, Howell JB, Shaia WT, Coelho DH (2020) Do postop-
17. Kim SY, Park JE, Lee YJ et al (2013) Testing a tool for assess- erative antibiotics improve skin reactivity following percutane-
ing the risk of bias for nonrandomized studies showed moderate ous auditory osseointegrated implant placement? Otol Neurotol
reliability and promising validity. J Clin Epidemiol 66:408–414. 20:802–805. https​://doi.org/10.1097/mao.00000​00000​00264​2
https​://doi.org/10.1016/j.jclin​epi.2012.09.016 33. di Yin G, Zeng X, Li P (2015) Skin reactions caused by bone-
18. Wallace BC, Schmid CH, Lau J, Trikalinos TA (2009) Meta- anchored hearing aid (BAHA) implantation. J Otol 10:159–162.
analyst: software for meta-analysis of binary, continuous and https​://doi.org/10.1016/j.joto.2016.01.001
diagnostic data. BMC Med Res Methodol 9:80. https​://doi. 34. Sardiwalla Y, Jufas N, Morris DP (2017) Direct cost compari-
org/10.1186/1471-2288-9-80 son of minimally invasive punch technique versus traditional
19. Wan X, Wang W, Liu J, Tong T (2014) Estimating the sample approaches for percutaneous bone anchored hearing devices. J
mean and standard deviation from the sample size, median, range Otolaryngol Head Neck Surg 46:46. https:​ //doi.org/10.1186/s4046​
and/or interquartile range. BMC Med Res Methodol 14:135 3-017-0222-2
20. Wilson DF, Kim HH (2013) A minimally invasive technique for 35. Wazen J, Smith J, Daugherty J (2019) Implantable auditory
the implantation of bone-anchored hearing devices. Otolaryngol devices: financial considerations and office-based implantation.
Head Neck Surg 149:473–477. https​://doi.org/10.1177/01945​ Otolaryngol Clin North Am 52:357–361
99813​49294​6
21. Dumon T, Wegner I, Sperling N, Grolman W (2017) Implanta- Publisher’s Note Springer Nature remains neutral with regard to
tion of bone-anchored hearing devices through a minimal skin jurisdictional claims in published maps and institutional affiliations.
punch incision versus the epidermal flap technique. Otol Neurotol
38:89–96. https​://doi.org/10.1097/MAO.00000​00000​00125​8
22. Bonilla A, Magri C, Juan E (2017) Findings from the experience
with the punch technique for auditory osseointegrated implants:

13