(218 days)
KLS Martin Individual Patient Solutions implant devices are intended for use in the stabilization, fixation, and reconstruction of the maxillofacial / midface and mandibular skeletal regions.
KLS Martin Individual Patient Solutions is comprised of patient-specific models and metallic bone plates used in conjunction with metallic bone screws for internal fixation of maxillofacial / midface and mandibular bones. The devices are manufactured based on medical imaging (CT scan) of the patient's anatomy with input from the physician during virtual planning and prior to finalization and production of the device. The physician only provides input for model manipulation and interactive feedback by viewing digital models of planned outputs that are modified by trained KLS Martin engineers during the planning session. For each design iteration, verification is performed by virtually fitting the generated implant over a 3D model of the patient's anatomy to ensure its dimensional properties allow an adequate fit.
Implants are provided non-sterile, range in thickness from 0.3 mm - 10 mm, and are manufactured using traditional (subtractive) methods from either CP Titanium (ASTM F67) or Ti-6Al-4V (ASTM F136) materials or additive methods from Ti-6Al-4V. These patient-specific devices are fixated with previously cleared KLS Martin screws.
The provided text describes the performance testing of the KLS Martin Individual Patient Solutions device, primarily focusing on non-clinical bench testing to demonstrate substantial equivalence to predicate devices, rather than a clinical study establishing device performance against acceptance criteria in human subjects.
Therefore, many of the requested details regarding clinical study design (e.g., sample size for test set, expert adjudication, MRMC study, standalone performance, ground truth establishment for training/test sets) are not applicable as they relate to clinical studies that were explicitly stated as "not necessary for the substantial equivalence determination."
However, I can extract information related to the non-clinical performance testing and the implicit acceptance criteria derived from comparison to predicate devices and established standards.
Here's a breakdown of the available information:
Acceptance Criteria and Reported Device Performance (Non-Clinical)
The acceptance criteria are generally implied to be meeting or exceeding the performance and safety profiles of the predicate devices and relevant ASTM/ISO standards. The "reported device performance" is framed as demonstrating substantial equivalence rather than specific numerical metrics for a clinical outcome.
| Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|
| Mechanical Performance (Tensile & Bending): Equivalent or superior bending resistance and fatigue life to predicate devices (specifically K944565) as per ASTM F382. | "The bending resistance and fatigue life of the subject devices was determined to be substantially equivalent to the K944565 plates." |
| Biocompatibility: Meet ISO 10993-1 standards and be equivalent to predicate (K163579) for titanium devices. | "Biocompatibility endpoints were evaluated in accordance with ISO 10993-1... The battery of cytotoxicity, chemical analysis, sensitization and irritation, and chemical/material characterization testing was leveraged from K163579 for titanium devices. The subject devices are identical to the primary predicate devices in material formulations, manufacturing methods and processes, and sterilization methods. No other chemicals have been added..." |
| Sterilization: Achieve a sterility assurance level (SAL) of 10-6 using the biological indicator (BI) overkill method as per ISO 17665-1:2006. | "Steam sterilization validations were performed... All test method acceptance criteria were met. Validations for devices manufactured from titanium were leveraged from the predicate device, KLS Martin Individual Patient Solutions (K163579). Subject titanium devices are identical in formulation, manufacturing processes, and post-processing procedures (cleaning & sterilization) as the predicate device." |
| Pyrogenicity: Contain endotoxin levels below the USP allowed limit for medical devices as per AAMI ANSI ST72. | "LAL endotoxin testing was conducted according to AAMI ANSI ST72... The results of the testing demonstrate that the subject devices contain endotoxin levels below the USP allowed limit for medical devices and meet pyrogen limit specifications. LAL endotoxin testing for titanium was leveraged from the predicate device, KLS Martin Individual Patient Solutions (K163579)." |
| Software Verification & Validation: All software requirements and specifications are implemented correctly and completely, traceable to system requirements, and conform to pre-defined specifications and acceptance criteria. | "Quality and on-site user acceptance testing provide objective evidence that all software requirements and specifications were implemented correctly and completely and are traceable to system requirements. Testing required as a result of risk analysis and impact assessments showed conformity with pre-defined specifications and acceptance criteria. Software documentation demonstrates all appropriate steps have been taken to ensure mitigation of any potential risks and performs as intended based on the user requirements and specifications." |
Study Information (Non-Clinical Focus)
-
Sample size used for the test set and the data provenance:
- Test Set Sample Size: Not explicitly stated in terms of number of unique devices for each mechanical test, but rather described as "bench testing" and "comparative performance testing." It implies sufficient samples were tested to meet standard requirements for ASTM F382 and other bench tests.
- Data Provenance: The tests were conducted to demonstrate substantial equivalence to predicate devices (K163579, K944565), and some performance data (biocompatibility, sterilization, pyrogenicity) were "leveraged" from previous clearances of the predicate device since the materials and processes are identical. This implies the data originates from the manufacturer's internal testing or contract labs. The country of origin for the data is not specified but is presumably within the regulatory framework acceptable to the FDA. The testing is non-clinical bench testing, not retrospective or prospective human data.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience):
- Not Applicable. This pertains to clinical studies involving expert review of diagnostic images or outcomes. The provided text describes non-clinical bench testing and software validation.
-
Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Not Applicable. This pertains to clinical studies involving human interpretation and ground truth establishment.
-
If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:
- Not Applicable. No clinical study (MRMC or otherwise) involving human readers or AI assistance was conducted or described for this device, as "Clinical testing was not necessary for the substantial equivalence determination." This device consists of patient-specific implants and the related planning system, not an AI for image interpretation.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Not Applicable. While software verification and validation were performed for the planning software, it's not an "algorithm" in the sense of an AI model for diagnosis. The software is a tool for design and planning, with human input from the physician and KLS Martin engineers. Therefore, a "standalone algorithm performance" as typically defined for AI/ML devices is not relevant here.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- For Bench Testing: The "ground truth" for non-clinical performance testing is established by validated test methods (e.g., ASTM, ISO standards) and comparison to the known performance of the predicate devices. For software, the ground truth is its pre-defined specifications and user requirements.
- Not Applicable for clinical "ground truth" types mentioned (expert consensus, pathology, outcomes data).
-
The sample size for the training set:
- Not Applicable. This refers to machine learning models. The device involves patient-specific design based on CT scans, but not a generalizable AI model that requires a training set in the conventional sense. The software's design and functionality are established through traditional software development and validation, not machine learning training.
-
How the ground truth for the training set was established:
- Not Applicable. (See point 7).
§ 872.4760 Bone plate.
(a)
Identification. A bone plate is a metal device intended to stabilize fractured bone structures in the oral cavity. The bone segments are attached to the plate with screws to prevent movement of the segments.(b)
Classification. Class II.