The primary efficacy variable is Survival through the late follow-up (LFU) visit. Survival is achieved when the participant is alive through the LFU visit. No other factors are considered in the evaluation of survival.

Modified intent-to-treat (mITT) population (included all subjects who had a culture-confirmed Gram-negative bacteria that had been treated with at least one dose of study treatment, and had an APACHE II score ≥ 10 at the time of diagnosis of pneumonia)

Death through LFU visit was adjudicated as pneumonia-related or pneumonia-unrelated for participants in the amikacin inhale group and participants in the placebo group.

Participants who died through LFU visit in mITT population set

Early Clinical Response was determined by the following: 1. CPIS scoring at Days 3, 5, and 10 compared to baseline (a. On Day 3, CPIS increase from baseline by at least 2 points was considered a failure. b. On Day 5, CPIS decrease from baseline of at least 1 point was not a failure. CPIS of no change from baseline was considered a failure. Any CPIS increase from baseline was a failure. c. On Day 10, CPIS decrease from baseline of at least 2 points was not a failure. CPIS decrease of only 1 point is a failure. Clinical Pulmonary Infection Score of no change was considered a failure. Any CPIS increase from baseline was a failure). 2. All-cause mortality through EOT visit was a failure. 3. The development of empyema or lung abscess through the EOT visit was a failure.

mITT population

Number of days on mechanical ventilator was summarized by descriptive statistics. Duration was defined as the number of days from the date of first study drug through the LFU visit. For participants who lived through the LFU visit, the ventilation days were actual days on ventilation with a maximum value of 28 days. For participants who died after Day 28 but on or before their LFU visit, the days on ventilator was censored at 28 days. For participants who died or discontinued off ventilation, the number of days on ventilation was actual days on ventilation with a maximum value of 28 days. For participants who died or discontinued on ventilation, the number of days on ventilation was 28 days. Further analysis of the number of days on mechanical ventilator was to be performed with censoring at Day 28 for subset of participants on ventilation without censoring.

mITT population

Number of days in ICU was summarized by descriptive statistics. Duration was defined as the number of days from the date of first study drug through the LFU visit. For participants who lived in ICU through the LFU visit, the ICU days were actual days in ICU with a maximum value of 28 days. For participants who died after Day 28 but on or before their LFU visit, the days in ICU was censored at 28 days. For participants who died or discontinued in ICU, the number of days in ICU was 28 days. Further analysis of the number of days in ICU was to be performed with censoring at Day 28 for subset of participants on ventilation and without censoring.

mITT population without missing start or end dates

The number of participants with microbiological response for each pathogen among the total number of participants with baseline pathogen isolates for each pathogen was determined. If a participant had 3 pathogens, all 3 were tabulated. Eradication ( defined as the absence of the original pathogen(s) at the post-treatment test-of-cure [TOC] visit culture of specimens from the original site of infection) and presumed eradication (defined as absence of appropriate culture material in a participant judged to be a clinical cure; he or she was unable to produce sputum and invasive procedures were not warranted) rates were reported to reveal the microbiological responses. The data were displayed for each bacterial genus/species. Baseline pathogen was defined as pathogens tested at Screening and Day 1 visit by central laboratory.

mITT population

The responses of eradication (defined as the absence of the original pathogen(s) at the post-treatment TOC culture of specimens from the original site of infection) and presumed eradication (defined as absence of appropriate culture material in a participant judged to be a clinical cure; he or she was unable to produce sputum and invasive procedures were not warranted) were tabulated for each participant to reveal the microbiological responses. All pathogen isolates from a participant must be eradicated (or presumed eradicated) to tabulate an eradicated (or presumed eradicated) response. Baseline pathogen was defined as pathogens tested at Screening and Day 1 visit by central laboratory.

mITT population

The responses of recurrence were tabulated for each participant. Recurrence was defined as the reappearance of the original pathogen(s) from a specimen taken after the TOC visit. If one or more pathogen reappeared, all isolates from a participant were tabulated as "recurrence". Baseline pathogen was defined as pathogens tested at Screening and Day 1 visit by central laboratory.

mITT population

New pathogens also denoted as superinfection was defined as the isolation of a new pathogen (not the original baseline pathogen) from a specimen taken while the participant was on antibiotic therapy (Day 1 to EOT) and having a need for alternative antimicrobial therapy. Rates of emergence of any new pathogen by participant after start of study drug were summarized for each treatment group.

mITT population

Resistance to amikacin was determined for the bacterial isolates by using a standardized microbiology laboratory test that generates a minimum inhibitory concentration (MIC) for amikacin and bacterial isolate. The same microbiology resistance standard was used for all bacteria tested against amikacin. Resistant bacteria have a MIC value of 64 μg/mL or greater. Percentages of resistance were calculated based on the percentage of participants infected with any treatment-emergent pathogens resistant to amikacin. If a participant had a more than one occurrence of a specific pathogen during pre-treatment period, the worst case of testing was used.

mITT population participants with pathogen susceptible to amikacin in pre-treatment period

AE was untoward medical occurrence in participant who received study drug without regard to possibility of causal relationship. AEs, occurred any time after the first dose of therapy and through 7 days after the EOT were recorded as treat-emergent AEs (TEAEs).

ITT for Safety population (included all participants in ITT analysis set who were analyzed as treated for safety analyses)

AE was untoward medical occurrence in participant who received study drug without regard to possibility of causal relationship. Serious AE: AE resulting in following outcomes or deemed significant for any reason: death; life-threatening; inpatient hospitalization or prolongation of existing hospitalization; persistent; significant disability/incapacity; congenital anomaly/birth defect; medical important serious event judged by investigator. SAEs, occurred any time after the first dose of therapy and through 7 days after the EOT were recorded as treat-emergent SAEs (TESAEs).

ITT for Safety population

The overall number of participants with any organ failure was summarized for each treatment group. Organ failure was defined by a specific organ type and by a collection of MedDRA version 20.0 preferred terms that were determined by the sponsor's clinical team. A participant with multiple AEs within a system organ class or preferred term is counted a single time for that system organ class (SOC) or preferred term.

ITT for Safety population

Number of deaths due to any reason through Day 10 and Day 15 were summarized for each treatment group.

mITT population