Table 2 Key findings from studies evaluating the role of acoustic analysis in detecting implant stability and subsidence

Goossens et al. (2020)

BPF quantifies the relative spectral power distribution of the measured sound signal. PCCs were calculated as a distance metric between the vibroacoustic response spectra of successive insertion hammer blows

18 of 23 implants showed a pattern of increasing BPF and PCCs. BPF best-detected implant seating. A 1 mm subsidence corresponded to approximately a doubling of the BPF value on consecutive hammer blows

Zhuang et al. (2022) [1, 18]

Hammering sounds were compared between hips with and without postoperative subsidence. The frequency spectrum was divided into 25 frequency bands at 0.5 kHz intervals. SP and nSP^a were calculated per frequency band

Average subsidence was 2.15 ± 2.91 mm. 9 acoustic parameters were significantly associated with the value of subsidence. The acoustic bands between 5 and 9.5 kHz had significantly lower amounts in hips with subsidence

The nSP was higher for those with subsidence in the 0.5 to 2.0 kHz range (divided by 0.5 intervals into 3 sub-groups, all significant)

Morohashi (2017) [3]

Pattern A = frequencies near 7 kHz became more accentuated as implantation progressed (n = 42)

Pattern B = no accentuation of frequencies near 7 kHz (n = 29)

 > 1 mm subsidence in 6 of 42 hips in patients with Pattern A and 11 of 27 hips in those with Pattern B, P = 0.013

Pastrav et al. (2009) [19]

The FRF changes were used to assess the evolution of the stiffness of the implant-bone structure and, consequently, the progression of the implant stability. When no noticeable change was observed in the FRF graph, hammering was stopped. The similarity of two successive FRF graphs was evaluated using PCCs. A correlation between the FRFs of consecutive stages of R = (0.99 ± 0.01) was considered as the endpoint

A FRF graph right shift indicates increased stiffness and normal evolution between successive hammer blows as reflected by increasing resonance frequencies. Additionally, the higher resonance frequencies are more sensitive to change in implant stability as opposed to the lower resonance frequencies. For non-cemented stems, the FRF between the 4th and 5th hammer blows were correlated > 0.99 for 26/30 cases (86.7%). The other 4/30 were > 0.95 correlation. Therefore, the PCCs between successive FRFs can be used as an endpoint criterion for insertion. For cemented stems, there was a notable difference in FRF between the non-cemented vs. cemented stage (after cement curing) in 85% of these cases (n = 45). The shift to the right of the graph indicates an increased stability after cementation. In the remaining 15% (n = 8), cement curing did not substantially change the FRF graphs, likely because the stems were already fixed when the cement polymerization was complete

Sakai et al. (2022) [15]

When the maximum peak frequency stays within the range of ± 0.5 kHz three times in a row, the stem was deemed fixed and the miniaturized analysis system provided a warning that further hammering would cause a fracture

In all 12 cases, the system successfully determined stem stability. There was no report of subsidence, implant failure, aseptic loosening, or fracture through a five-year follow-up. The frequency determined at the time of attainment of implant stability was 4.02 ± 2.33 kHz. There were no reports of subsidence in the immediate postoperative period

Sakai et al. (2021) [16]

Stability and cup fixation were defined when the maximum peak frequency changed within ± 0.5 kHz or less in three consecutive blows

The mean stable maximum peak frequency was 4.42 ± 4.02 kHz. A constant maximum peak frequency continued 3.27 ± 0.47 hammering counts. Once the maximum peak frequency stabilizes, as represented by maximum peak frequency ± 0.5 kHz on consecutive blows, hammering should be stopped. There were no cases of immediate postoperative subsidence

Homma et al. (2022) [7]

During final size broaching, the 5 hammering sounds before the last blow were included for analysis

12 out of 62 (19.4%) showed ≥ 3 mm of post-operative subsidence

Homma et al. (2023) [2, 20]

The first three hammering sounds and the final hammering sound were not included to avoid any inconsistencies in hammer impactions. The early phase was defined as the 4th–6th hammering sounds from the beginning. The late phase was defined as the 2nd–4th hammering sounds from the end stem insertion. The frequency spectrum of the hammering sounds was divided into 25 frequency bands at 0.5 kHz intervals. The SP and NSP were assessed for each frequency band. Alternation ratio^b was used to define the change in acoustic characteristics

Acoustic characteristics were significantly different between the late phase and the early phase. There was an augmentation of low-frequency bands, as shown by the different alteration ratios of the low-frequency bands (0.5–1 kHz and 1 kHz). For analysis of sound alterations, the low-frequency bands (0.5–1.0 kHz and 1.0–1.5 kHz) were considered the key bands as they showed the most considerable changes

Mulier et al. (2008) [4]

The amount of FRF change between insertion steps was evaluated by calculating PCCs between successive FRFs. A correlation between the FRFs of consecutive stages of R = (0.99 ± 0.01) was considered as the endpoint

In 26 cases (86.7%), the correlation coefficient between the last two FRFs was above 0.99. In the other four cases (13.3%), the surgeon ceased the insertion because of suspected bone fragility

McConnell et al. (2018) [17]

Detection of a low-frequency band centered around 1036 Hz (IQR 944 to 1093) during the final femoral broach represented a ‘change in sound’. This band had not been detected in any of the prior broach impactions

In 75 hips, a "change in sound" was associated with a well-fitted stem (true positive). In 2 hips, a "change in sound" was detected prior to final broaching (false positive). In 24 hips, no change in sound was detected. Out of these 24 hips, 9 hips were judged to have an undersized implant (true negative). The remaining 15 hips in which no change in sound was detected had a well-fitted prosthesis (false negative)